| 1 |
"""Tests for HTMLParser.py.""" |
| 2 |
|
| 3 |
import html.parser |
| 4 |
import pprint |
| 5 |
import unittest |
| 6 |
|
| 7 |
from unittest.mock import patch |
| 8 |
from test import support |
| 9 |
|
| 10 |
|
| 11 |
SAMPLE_RCDATA = ( |
| 12 |
'<!-- not a comment -->' |
| 13 |
"<not a='start tag'>" |
| 14 |
'<![CDATA[not a cdata]]>' |
| 15 |
'<!not a bogus comment>' |
| 16 |
'</not a bogus comment>' |
| 17 |
'\u2603' |
| 18 |
) |
| 19 |
|
| 20 |
SAMPLE_RAWTEXT = SAMPLE_RCDATA + '&☺' |
| 21 |
|
| 22 |
|
| 23 |
class EventCollector(html.parser.HTMLParser): |
| 24 |
|
| 25 |
def __init__(self, *args, autocdata=False, **kw): |
| 26 |
self.autocdata = autocdata |
| 27 |
self.events = [] |
| 28 |
self.append = self.events.append |
| 29 |
html.parser.HTMLParser.__init__(self, *args, **kw) |
| 30 |
if autocdata: |
| 31 |
self._set_support_cdata(False) |
| 32 |
|
| 33 |
def get_events(self): |
| 34 |
# Normalize the list of events so that buffer artefacts don't |
| 35 |
# separate runs of contiguous characters. |
| 36 |
L = [] |
| 37 |
prevtype = None |
| 38 |
for event in self.events: |
| 39 |
type = event[0] |
| 40 |
if type == prevtype == "data": |
| 41 |
L[-1] = ("data", L[-1][1] + event[1]) |
| 42 |
else: |
| 43 |
L.append(event) |
| 44 |
prevtype = type |
| 45 |
self.events = L |
| 46 |
return L |
| 47 |
|
| 48 |
# structure markup |
| 49 |
|
| 50 |
def handle_starttag(self, tag, attrs): |
| 51 |
self.append(("starttag", tag, attrs)) |
| 52 |
if self.autocdata and tag == 'svg': |
| 53 |
self._set_support_cdata(True) |
| 54 |
|
| 55 |
def handle_startendtag(self, tag, attrs): |
| 56 |
self.append(("startendtag", tag, attrs)) |
| 57 |
|
| 58 |
def handle_endtag(self, tag): |
| 59 |
self.append(("endtag", tag)) |
| 60 |
if self.autocdata and tag == 'svg': |
| 61 |
self._set_support_cdata(False) |
| 62 |
|
| 63 |
# all other markup |
| 64 |
|
| 65 |
def handle_comment(self, data): |
| 66 |
self.append(("comment", data)) |
| 67 |
|
| 68 |
def handle_charref(self, data): |
| 69 |
self.append(("charref", data)) |
| 70 |
|
| 71 |
def handle_data(self, data): |
| 72 |
self.append(("data", data)) |
| 73 |
|
| 74 |
def handle_decl(self, data): |
| 75 |
self.append(("decl", data)) |
| 76 |
|
| 77 |
def handle_entityref(self, data): |
| 78 |
self.append(("entityref", data)) |
| 79 |
|
| 80 |
def handle_pi(self, data): |
| 81 |
self.append(("pi", data)) |
| 82 |
|
| 83 |
def unknown_decl(self, decl): |
| 84 |
self.append(("unknown decl", decl)) |
| 85 |
|
| 86 |
|
| 87 |
class EventCollectorExtra(EventCollector): |
| 88 |
|
| 89 |
def handle_starttag(self, tag, attrs): |
| 90 |
EventCollector.handle_starttag(self, tag, attrs) |
| 91 |
self.append(("starttag_text", self.get_starttag_text())) |
| 92 |
|
| 93 |
|
| 94 |
class EventCollectorCharrefs(EventCollector): |
| 95 |
|
| 96 |
def handle_charref(self, data): |
| 97 |
self.fail('This should never be called with convert_charrefs=True') |
| 98 |
|
| 99 |
def handle_entityref(self, data): |
| 100 |
self.fail('This should never be called with convert_charrefs=True') |
| 101 |
|
| 102 |
|
| 103 |
# The normal event collector normalizes the events in get_events, |
| 104 |
# so we override it to return the original list of events. |
| 105 |
class EventCollectorNoNormalize(EventCollector): |
| 106 |
def get_events(self): |
| 107 |
return self.events |
| 108 |
|
| 109 |
|
| 110 |
class TestCaseBase(unittest.TestCase): |
| 111 |
|
| 112 |
def get_collector(self, convert_charrefs=False): |
| 113 |
return EventCollector(convert_charrefs=convert_charrefs) |
| 114 |
|
| 115 |
def _run_check(self, source, expected_events, |
| 116 |
*, collector=None, convert_charrefs=False): |
| 117 |
if collector is None: |
| 118 |
collector = self.get_collector(convert_charrefs=convert_charrefs) |
| 119 |
parser = collector |
| 120 |
for s in source: |
| 121 |
parser.feed(s) |
| 122 |
parser.close() |
| 123 |
events = parser.get_events() |
| 124 |
if events != expected_events: |
| 125 |
self.fail("received events did not match expected events" + |
| 126 |
"\nSource:\n" + repr(source) + |
| 127 |
"\nExpected:\n" + pprint.pformat(expected_events) + |
| 128 |
"\nReceived:\n" + pprint.pformat(events)) |
| 129 |
|
| 130 |
def _run_check_extra(self, source, events): |
| 131 |
self._run_check(source, events, |
| 132 |
collector=EventCollectorExtra(convert_charrefs=False)) |
| 133 |
|
| 134 |
|
| 135 |
class HTMLParserTestCase(TestCaseBase): |
| 136 |
|
| 137 |
def test_processing_instruction_only(self): |
| 138 |
self._run_check("<?processing instruction>", [ |
| 139 |
("pi", "processing instruction"), |
| 140 |
]) |
| 141 |
self._run_check("<?processing instruction ?>", [ |
| 142 |
("pi", "processing instruction ?"), |
| 143 |
]) |
| 144 |
|
| 145 |
def test_simple_html(self): |
| 146 |
self._run_check(""" |
| 147 |
<!DOCTYPE html PUBLIC 'foo'> |
| 148 |
<HTML>&entity;  |
| 149 |
<!--comment1a |
| 150 |
-></foo><bar><<?pi?></foo<bar |
| 151 |
comment1b--> |
| 152 |
<Img sRc='Bar' isMAP>sample |
| 153 |
text |
| 154 |
“ |
| 155 |
<!--comment2a-- --comment2b--> |
| 156 |
</Html> |
| 157 |
""", [ |
| 158 |
("data", "\n"), |
| 159 |
("decl", "DOCTYPE html PUBLIC 'foo'"), |
| 160 |
("data", "\n"), |
| 161 |
("starttag", "html", []), |
| 162 |
("entityref", "entity"), |
| 163 |
("charref", "32"), |
| 164 |
("data", "\n"), |
| 165 |
("comment", "comment1a\n-></foo><bar><<?pi?></foo<bar\ncomment1b"), |
| 166 |
("data", "\n"), |
| 167 |
("starttag", "img", [("src", "Bar"), ("ismap", None)]), |
| 168 |
("data", "sample\ntext\n"), |
| 169 |
("charref", "x201C"), |
| 170 |
("data", "\n"), |
| 171 |
("comment", "comment2a-- --comment2b"), |
| 172 |
("data", "\n"), |
| 173 |
("endtag", "html"), |
| 174 |
("data", "\n"), |
| 175 |
]) |
| 176 |
|
| 177 |
def test_malformatted_charref(self): |
| 178 |
self._run_check("<p>&#bad;</p>", [ |
| 179 |
("starttag", "p", []), |
| 180 |
("data", "&#bad;"), |
| 181 |
("endtag", "p"), |
| 182 |
]) |
| 183 |
# add the [] as a workaround to avoid buffering (see #20288) |
| 184 |
self._run_check(["<div>&#bad;</div>"], [ |
| 185 |
("starttag", "div", []), |
| 186 |
("data", "&#bad;"), |
| 187 |
("endtag", "div"), |
| 188 |
]) |
| 189 |
|
| 190 |
def test_unclosed_entityref(self): |
| 191 |
self._run_check('> <', [('entityref', 'gt'), ('data', ' '), ('entityref', 'lt')], |
| 192 |
convert_charrefs=False) |
| 193 |
self._run_check('> <', [('data', '> <')], convert_charrefs=True) |
| 194 |
|
| 195 |
self._run_check('&undefined <', |
| 196 |
[('entityref', 'undefined'), ('data', ' '), ('entityref', 'lt')], |
| 197 |
convert_charrefs=False) |
| 198 |
self._run_check('&undefined <', [('data', '&undefined <')], |
| 199 |
convert_charrefs=True) |
| 200 |
|
| 201 |
self._run_check('>undefined <', |
| 202 |
[('entityref', 'gtundefined'), ('data', ' '), ('entityref', 'lt')], |
| 203 |
convert_charrefs=False) |
| 204 |
self._run_check('>undefined <', [('data', '>undefined <')], |
| 205 |
convert_charrefs=True) |
| 206 |
|
| 207 |
self._run_check('& <', [('data', '& '), ('entityref', 'lt')], |
| 208 |
convert_charrefs=False) |
| 209 |
self._run_check('& <', [('data', '& <')], convert_charrefs=True) |
| 210 |
|
| 211 |
def test_eof_in_entityref(self): |
| 212 |
self._run_check('>', [('entityref', 'gt')], convert_charrefs=False) |
| 213 |
self._run_check('>', [('data', '>')], convert_charrefs=True) |
| 214 |
|
| 215 |
self._run_check('&g', [('entityref', 'g')], convert_charrefs=False) |
| 216 |
self._run_check('&g', [('data', '&g')], convert_charrefs=True) |
| 217 |
|
| 218 |
self._run_check('&undefined', [('entityref', 'undefined')], |
| 219 |
convert_charrefs=False) |
| 220 |
self._run_check('&undefined', [('data', '&undefined')], |
| 221 |
convert_charrefs=True) |
| 222 |
|
| 223 |
self._run_check('>undefined', [('entityref', 'gtundefined')], |
| 224 |
convert_charrefs=False) |
| 225 |
self._run_check('>undefined', [('data', '>undefined')], |
| 226 |
convert_charrefs=True) |
| 227 |
|
| 228 |
self._run_check('&', [('data', '&')], convert_charrefs=False) |
| 229 |
self._run_check('&', [('data', '&')], convert_charrefs=True) |
| 230 |
|
| 231 |
def test_unclosed_charref(self): |
| 232 |
self._run_check('{ <', [('charref', '123'), ('data', ' '), ('entityref', 'lt')], |
| 233 |
convert_charrefs=False) |
| 234 |
self._run_check('{ <', [('data', '{ <')], convert_charrefs=True) |
| 235 |
self._run_check('« <', [('charref', 'xab'), ('data', ' '), ('entityref', 'lt')], |
| 236 |
convert_charrefs=False) |
| 237 |
self._run_check('« <', [('data', '\xab <')], convert_charrefs=True) |
| 238 |
|
| 239 |
self._run_check('� <', |
| 240 |
[('charref', '123456789'), ('data', ' '), ('entityref', 'lt')], |
| 241 |
convert_charrefs=False) |
| 242 |
self._run_check('� <', [('data', '\ufffd <')], |
| 243 |
convert_charrefs=True) |
| 244 |
self._run_check('� <', |
| 245 |
[('charref', 'x123456789'), ('data', ' '), ('entityref', 'lt')], |
| 246 |
convert_charrefs=False) |
| 247 |
self._run_check('� <', [('data', '\ufffd <')], |
| 248 |
convert_charrefs=True) |
| 249 |
|
| 250 |
self._run_check('&# <', [('data', '&# '), ('entityref', 'lt')], convert_charrefs=False) |
| 251 |
self._run_check('&# <', [('data', '&# <')], convert_charrefs=True) |
| 252 |
self._run_check('&#x <', [('data', '&#x '), ('entityref', 'lt')], convert_charrefs=False) |
| 253 |
self._run_check('&#x <', [('data', '&#x <')], convert_charrefs=True) |
| 254 |
|
| 255 |
def test_eof_in_charref(self): |
| 256 |
self._run_check('{', [('charref', '123')], convert_charrefs=False) |
| 257 |
self._run_check('{', [('data', '{')], convert_charrefs=True) |
| 258 |
self._run_check('«', [('charref', 'xab')], convert_charrefs=False) |
| 259 |
self._run_check('«', [('data', '\xab')], convert_charrefs=True) |
| 260 |
|
| 261 |
self._run_check('�', [('charref', '123456789')], |
| 262 |
convert_charrefs=False) |
| 263 |
self._run_check('�', [('data', '\ufffd')], convert_charrefs=True) |
| 264 |
self._run_check('�', [('charref', 'x123456789')], |
| 265 |
convert_charrefs=False) |
| 266 |
self._run_check('�', [('data', '\ufffd')], convert_charrefs=True) |
| 267 |
|
| 268 |
self._run_check('&#', [('data', '&#')], convert_charrefs=False) |
| 269 |
self._run_check('&#', [('data', '&#')], convert_charrefs=True) |
| 270 |
self._run_check('&#x', [('data', '&#x')], convert_charrefs=False) |
| 271 |
self._run_check('&#x', [('data', '&#x')], convert_charrefs=True) |
| 272 |
|
| 273 |
def test_bad_nesting(self): |
| 274 |
# Strangely, this *is* supposed to test that overlapping |
| 275 |
# elements are allowed. HTMLParser is more geared toward |
| 276 |
# lexing the input that parsing the structure. |
| 277 |
self._run_check("<a><b></a></b>", [ |
| 278 |
("starttag", "a", []), |
| 279 |
("starttag", "b", []), |
| 280 |
("endtag", "a"), |
| 281 |
("endtag", "b"), |
| 282 |
]) |
| 283 |
|
| 284 |
def test_bare_ampersands(self): |
| 285 |
self._run_check("this text & contains & ampersands &", [ |
| 286 |
("data", "this text & contains & ampersands &"), |
| 287 |
]) |
| 288 |
|
| 289 |
def test_bare_pointy_brackets(self): |
| 290 |
self._run_check("this < text > contains < bare>pointy< brackets", [ |
| 291 |
("data", "this < text > contains < bare>pointy< brackets"), |
| 292 |
]) |
| 293 |
|
| 294 |
def test_starttag_end_boundary(self): |
| 295 |
self._run_check("""<a b='<'>""", [("starttag", "a", [("b", "<")])]) |
| 296 |
self._run_check("""<a b='>'>""", [("starttag", "a", [("b", ">")])]) |
| 297 |
|
| 298 |
def test_buffer_artefacts(self): |
| 299 |
output = [("starttag", "a", [("b", "<")])] |
| 300 |
self._run_check(["<a b='<'>"], output) |
| 301 |
self._run_check(["<a ", "b='<'>"], output) |
| 302 |
self._run_check(["<a b", "='<'>"], output) |
| 303 |
self._run_check(["<a b=", "'<'>"], output) |
| 304 |
self._run_check(["<a b='<", "'>"], output) |
| 305 |
self._run_check(["<a b='<'", ">"], output) |
| 306 |
|
| 307 |
output = [("starttag", "a", [("b", ">")])] |
| 308 |
self._run_check(["<a b='>'>"], output) |
| 309 |
self._run_check(["<a ", "b='>'>"], output) |
| 310 |
self._run_check(["<a b", "='>'>"], output) |
| 311 |
self._run_check(["<a b=", "'>'>"], output) |
| 312 |
self._run_check(["<a b='>", "'>"], output) |
| 313 |
self._run_check(["<a b='>'", ">"], output) |
| 314 |
|
| 315 |
output = [("comment", "abc")] |
| 316 |
self._run_check(["", "<!--abc-->"], output) |
| 317 |
self._run_check(["<", "!--abc-->"], output) |
| 318 |
self._run_check(["<!", "--abc-->"], output) |
| 319 |
self._run_check(["<!-", "-abc-->"], output) |
| 320 |
self._run_check(["<!--", "abc-->"], output) |
| 321 |
self._run_check(["<!--a", "bc-->"], output) |
| 322 |
self._run_check(["<!--ab", "c-->"], output) |
| 323 |
self._run_check(["<!--abc", "-->"], output) |
| 324 |
self._run_check(["<!--abc-", "->"], output) |
| 325 |
self._run_check(["<!--abc--", ">"], output) |
| 326 |
self._run_check(["<!--abc-->", ""], output) |
| 327 |
|
| 328 |
def test_valid_doctypes(self): |
| 329 |
# from http://www.w3.org/QA/2002/04/valid-dtd-list.html |
| 330 |
dtds = ['HTML', # HTML5 doctype |
| 331 |
('HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" ' |
| 332 |
'"http://www.w3.org/TR/html4/strict.dtd"'), |
| 333 |
('HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" ' |
| 334 |
'"http://www.w3.org/TR/html4/loose.dtd"'), |
| 335 |
('html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" ' |
| 336 |
'"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"'), |
| 337 |
('html PUBLIC "-//W3C//DTD XHTML 1.0 Frameset//EN" ' |
| 338 |
'"http://www.w3.org/TR/xhtml1/DTD/xhtml1-frameset.dtd"'), |
| 339 |
('math PUBLIC "-//W3C//DTD MathML 2.0//EN" ' |
| 340 |
'"http://www.w3.org/Math/DTD/mathml2/mathml2.dtd"'), |
| 341 |
('html PUBLIC "-//W3C//DTD ' |
| 342 |
'XHTML 1.1 plus MathML 2.0 plus SVG 1.1//EN" ' |
| 343 |
'"http://www.w3.org/2002/04/xhtml-math-svg/xhtml-math-svg.dtd"'), |
| 344 |
('svg PUBLIC "-//W3C//DTD SVG 1.1//EN" ' |
| 345 |
'"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd"'), |
| 346 |
'html PUBLIC "-//IETF//DTD HTML 2.0//EN"', |
| 347 |
'html PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"'] |
| 348 |
for dtd in dtds: |
| 349 |
self._run_check("<!DOCTYPE %s>" % dtd, |
| 350 |
[('decl', 'DOCTYPE ' + dtd)]) |
| 351 |
|
| 352 |
def test_startendtag(self): |
| 353 |
self._run_check("<p/>", [ |
| 354 |
("startendtag", "p", []), |
| 355 |
]) |
| 356 |
self._run_check("<p></p>", [ |
| 357 |
("starttag", "p", []), |
| 358 |
("endtag", "p"), |
| 359 |
]) |
| 360 |
self._run_check("<p><img src='foo' /></p>", [ |
| 361 |
("starttag", "p", []), |
| 362 |
("startendtag", "img", [("src", "foo")]), |
| 363 |
("endtag", "p"), |
| 364 |
]) |
| 365 |
|
| 366 |
def test_get_starttag_text(self): |
| 367 |
s = """<foo:bar \n one="1"\ttwo=2 >""" |
| 368 |
self._run_check_extra(s, [ |
| 369 |
("starttag", "foo:bar", [("one", "1"), ("two", "2")]), |
| 370 |
("starttag_text", s)]) |
| 371 |
|
| 372 |
@support.subTests('content', [ |
| 373 |
'<!-- not a comment --> ¬-an-entity-ref;', |
| 374 |
"<not a='start tag'>", |
| 375 |
'<a href="" /> <p> <span></span>', |
| 376 |
'foo = "</scr" + "ipt>";', |
| 377 |
'foo = "</SCRIPT" + ">";', |
| 378 |
'foo = <\n/script> ', |
| 379 |
'<!-- document.write("</scr" + "ipt>"); -->', |
| 380 |
('\n//<![CDATA[\n' |
| 381 |
'document.write(\'<s\'+\'cript type="text/javascript" ' |
| 382 |
'src="http://www.example.org/r=\'+new ' |
| 383 |
'Date().getTime()+\'"><\\/s\'+\'cript>\');\n//]]>'), |
| 384 |
'\n<!-- //\nvar foo = 3.14;\n// -->\n', |
| 385 |
'<!-- \u2603 -->', |
| 386 |
]) |
| 387 |
def test_script_content(self, content): |
| 388 |
s = f'<script>{content}</script>' |
| 389 |
self._run_check(s, [ |
| 390 |
("starttag", "script", []), |
| 391 |
("data", content), |
| 392 |
("endtag", "script"), |
| 393 |
]) |
| 394 |
|
| 395 |
@support.subTests('content', [ |
| 396 |
'a::before { content: "<!-- not a comment -->"; }', |
| 397 |
'a::before { content: "¬-an-entity-ref;"; }', |
| 398 |
'a::before { content: "<not a=\'start tag\'>"; }', |
| 399 |
'a::before { content: "\u2603"; }', |
| 400 |
]) |
| 401 |
def test_style_content(self, content): |
| 402 |
s = f'<style>{content}</style>' |
| 403 |
self._run_check(s, [("starttag", "style", []), |
| 404 |
("data", content), |
| 405 |
("endtag", "style")]) |
| 406 |
|
| 407 |
@support.subTests('tag', ['title', 'textarea']) |
| 408 |
def test_rcdata_content(self, tag): |
| 409 |
source = f"<{tag}>{SAMPLE_RCDATA}</{tag}>" |
| 410 |
self._run_check(source, [ |
| 411 |
("starttag", tag, []), |
| 412 |
("data", SAMPLE_RCDATA), |
| 413 |
("endtag", tag), |
| 414 |
]) |
| 415 |
source = f"<{tag}>&</{tag}>" |
| 416 |
self._run_check(source, [ |
| 417 |
("starttag", tag, []), |
| 418 |
('entityref', 'amp'), |
| 419 |
("endtag", tag), |
| 420 |
]) |
| 421 |
|
| 422 |
@support.subTests('tag', |
| 423 |
['style', 'xmp', 'iframe', 'noembed', 'noframes', 'script']) |
| 424 |
def test_rawtext_content(self, tag): |
| 425 |
source = f"<{tag}>{SAMPLE_RAWTEXT}</{tag}>" |
| 426 |
self._run_check(source, [ |
| 427 |
("starttag", tag, []), |
| 428 |
("data", SAMPLE_RAWTEXT), |
| 429 |
("endtag", tag), |
| 430 |
]) |
| 431 |
|
| 432 |
def test_noscript_content(self): |
| 433 |
source = f"<noscript>{SAMPLE_RAWTEXT}</noscript>" |
| 434 |
# scripting=False -- normal mode |
| 435 |
self._run_check(source, [ |
| 436 |
('starttag', 'noscript', []), |
| 437 |
('comment', ' not a comment '), |
| 438 |
('starttag', 'not', [('a', 'start tag')]), |
| 439 |
('unknown decl', 'CDATA[not a cdata'), |
| 440 |
('comment', 'not a bogus comment'), |
| 441 |
('endtag', 'not'), |
| 442 |
('data', '☃'), |
| 443 |
('entityref', 'amp'), |
| 444 |
('charref', '9786'), |
| 445 |
('endtag', 'noscript'), |
| 446 |
]) |
| 447 |
# scripting=True -- RAWTEXT mode |
| 448 |
self._run_check(source, [ |
| 449 |
("starttag", "noscript", []), |
| 450 |
("data", SAMPLE_RAWTEXT), |
| 451 |
("endtag", "noscript"), |
| 452 |
], collector=EventCollector(scripting=True)) |
| 453 |
|
| 454 |
def test_plaintext_content(self): |
| 455 |
content = SAMPLE_RAWTEXT + '</plaintext>' # not closing |
| 456 |
source = f"<plaintext>{content}" |
| 457 |
self._run_check(source, [ |
| 458 |
("starttag", "plaintext", []), |
| 459 |
("data", content), |
| 460 |
]) |
| 461 |
|
| 462 |
@support.subTests('endtag', ['script', 'SCRIPT', 'script ', 'script\n', |
| 463 |
'script/', 'script foo=bar', 'script foo=">"']) |
| 464 |
def test_script_closing_tag(self, endtag): |
| 465 |
# see issue #13358 |
| 466 |
# make sure that HTMLParser calls handle_data only once for each CDATA. |
| 467 |
content = """<!-- not a comment --> ¬-an-entity-ref; |
| 468 |
<a href="" /> </p><p> <span></span></style> |
| 469 |
'</script' + '>'""" |
| 470 |
s = f'<ScrIPt>{content}</{endtag}>' |
| 471 |
self._run_check(s, [("starttag", "script", []), |
| 472 |
("data", content), |
| 473 |
("endtag", "script")], |
| 474 |
collector=EventCollectorNoNormalize(convert_charrefs=False)) |
| 475 |
|
| 476 |
@support.subTests('tag', [ |
| 477 |
'script', 'style', 'xmp', 'iframe', 'noembed', 'noframes', |
| 478 |
'textarea', 'title', 'noscript', |
| 479 |
]) |
| 480 |
def test_closing_tag(self, tag): |
| 481 |
for endtag in [tag, tag.upper(), f'{tag} ', f'{tag}\n', |
| 482 |
f'{tag}/', f'{tag} foo=bar', f'{tag} foo=">"']: |
| 483 |
content = "<!-- not a comment --><i>Spam</i>" |
| 484 |
s = f'<{tag.upper()}>{content}</{endtag}>' |
| 485 |
self._run_check(s, [ |
| 486 |
("starttag", tag, []), |
| 487 |
('data', content), |
| 488 |
("endtag", tag), |
| 489 |
], collector=EventCollectorNoNormalize(convert_charrefs=False, scripting=True)) |
| 490 |
|
| 491 |
@support.subTests('tag', [ |
| 492 |
'script', 'style', 'xmp', 'iframe', 'noembed', 'noframes', |
| 493 |
'textarea', 'title', 'noscript', |
| 494 |
]) |
| 495 |
def test_invalid_closing_tag(self, tag): |
| 496 |
content = ( |
| 497 |
f'< /{tag}>' |
| 498 |
f'</ {tag}>' |
| 499 |
f'</{tag}x>' |
| 500 |
f'</{tag}\v>' |
| 501 |
f'</{tag}\xa0>' |
| 502 |
) |
| 503 |
source = f"<{tag}>{content}</{tag}>" |
| 504 |
self._run_check(source, [ |
| 505 |
("starttag", tag, []), |
| 506 |
("data", content), |
| 507 |
("endtag", tag), |
| 508 |
], collector=EventCollector(convert_charrefs=False, scripting=True)) |
| 509 |
|
| 510 |
@support.subTests('tag,endtag', [ |
| 511 |
('title', 'tıtle'), |
| 512 |
('style', 'ſtyle'), |
| 513 |
('style', 'ſtyle'), |
| 514 |
('style', 'style'), |
| 515 |
('iframe', 'ıframe'), |
| 516 |
('noframes', 'noframeſ'), |
| 517 |
('noscript', 'noſcript'), |
| 518 |
('noscript', 'noscrıpt'), |
| 519 |
('script', 'ſcript'), |
| 520 |
('script', 'scrıpt'), |
| 521 |
]) |
| 522 |
def test_invalid_nonascii_closing_tag(self, tag, endtag): |
| 523 |
content = f"<br></{endtag}>" |
| 524 |
source = f"<{tag}>{content}" |
| 525 |
self._run_check(source, [ |
| 526 |
("starttag", tag, []), |
| 527 |
("data", content), |
| 528 |
], collector=EventCollector(convert_charrefs=False, scripting=True)) |
| 529 |
source = f"<{tag}>{content}</{tag}>" |
| 530 |
self._run_check(source, [ |
| 531 |
("starttag", tag, []), |
| 532 |
("data", content), |
| 533 |
("endtag", tag), |
| 534 |
], collector=EventCollector(convert_charrefs=False, scripting=True)) |
| 535 |
|
| 536 |
@support.subTests('tail,end', [ |
| 537 |
('', False), |
| 538 |
('<', False), |
| 539 |
('</', False), |
| 540 |
('</s', False), |
| 541 |
('</script', False), |
| 542 |
('</script ', True), |
| 543 |
('</script foo=bar', True), |
| 544 |
('</script foo=">', True), |
| 545 |
]) |
| 546 |
def test_eof_in_script(self, tail, end): |
| 547 |
content = "a = 123" |
| 548 |
s = f'<ScrIPt>{content}{tail}' |
| 549 |
self._run_check(s, [("starttag", "script", []), |
| 550 |
("data", content if end else content + tail)], |
| 551 |
collector=EventCollectorNoNormalize(convert_charrefs=False)) |
| 552 |
|
| 553 |
@support.subTests('tail,end', [ |
| 554 |
('', False), |
| 555 |
('<', False), |
| 556 |
('</', False), |
| 557 |
('</t', False), |
| 558 |
('</title', False), |
| 559 |
('</title ', True), |
| 560 |
('</title foo=bar', True), |
| 561 |
('</title foo=">', True), |
| 562 |
]) |
| 563 |
def test_eof_in_title(self, tail, end): |
| 564 |
s = f'<TitLe>Egg & Spam{tail}' |
| 565 |
self._run_check(s, [("starttag", "title", []), |
| 566 |
("data", "Egg & Spam" + ('' if end else tail))], |
| 567 |
collector=EventCollectorNoNormalize(convert_charrefs=True)) |
| 568 |
self._run_check(s, [("starttag", "title", []), |
| 569 |
('data', 'Egg '), |
| 570 |
('entityref', 'amp'), |
| 571 |
('data', ' Spam' + ('' if end else tail))], |
| 572 |
collector=EventCollectorNoNormalize(convert_charrefs=False)) |
| 573 |
|
| 574 |
def test_comments(self): |
| 575 |
html = ("<!-- I'm a valid comment -->" |
| 576 |
'<!--me too!-->' |
| 577 |
'<!------>' |
| 578 |
'<!----->' |
| 579 |
'<!---->' |
| 580 |
# abrupt-closing-of-empty-comment |
| 581 |
'<!--->' |
| 582 |
'<!-->' |
| 583 |
'<!----I have many hyphens---->' |
| 584 |
'<!-- I have a > in the middle -->' |
| 585 |
'<!-- and I have -- in the middle! -->' |
| 586 |
'<!--incorrectly-closed-comment--!>' |
| 587 |
'<!----!>' |
| 588 |
'<!----!-->' |
| 589 |
'<!---- >-->' |
| 590 |
'<!---!>-->' |
| 591 |
'<!--!>-->' |
| 592 |
# nested-comment |
| 593 |
'<!-- <!-- nested --> -->' |
| 594 |
'<!--<!-->' |
| 595 |
'<!--<!--!>' |
| 596 |
) |
| 597 |
expected = [('comment', " I'm a valid comment "), |
| 598 |
('comment', 'me too!'), |
| 599 |
('comment', '--'), |
| 600 |
('comment', '-'), |
| 601 |
('comment', ''), |
| 602 |
('comment', ''), |
| 603 |
('comment', ''), |
| 604 |
('comment', '--I have many hyphens--'), |
| 605 |
('comment', ' I have a > in the middle '), |
| 606 |
('comment', ' and I have -- in the middle! '), |
| 607 |
('comment', 'incorrectly-closed-comment'), |
| 608 |
('comment', ''), |
| 609 |
('comment', '--!'), |
| 610 |
('comment', '-- >'), |
| 611 |
('comment', '-!>'), |
| 612 |
('comment', '!>'), |
| 613 |
('comment', ' <!-- nested '), ('data', ' -->'), |
| 614 |
('comment', '<!'), |
| 615 |
('comment', '<!'), |
| 616 |
] |
| 617 |
self._run_check(html, expected) |
| 618 |
|
| 619 |
def test_condcoms(self): |
| 620 |
html = ('<!--[if IE & !(lte IE 8)]>aren\'t<![endif]-->' |
| 621 |
'<!--[if IE 8]>condcoms<![endif]-->' |
| 622 |
'<!--[if lte IE 7]>pretty?<![endif]-->') |
| 623 |
expected = [('comment', "[if IE & !(lte IE 8)]>aren't<![endif]"), |
| 624 |
('comment', '[if IE 8]>condcoms<![endif]'), |
| 625 |
('comment', '[if lte IE 7]>pretty?<![endif]')] |
| 626 |
self._run_check(html, expected) |
| 627 |
|
| 628 |
def test_convert_charrefs(self): |
| 629 |
# default value for convert_charrefs is now True |
| 630 |
collector = lambda: EventCollectorCharrefs() |
| 631 |
self.assertTrue(collector().convert_charrefs) |
| 632 |
charrefs = ['"', '"', '"', '"', '"', '"'] |
| 633 |
# check charrefs in the middle of the text |
| 634 |
expected = [('starttag', 'a', []), ('data', 'a"z'), ('endtag', 'a')] |
| 635 |
for charref in charrefs: |
| 636 |
self._run_check('<a>a{0}z</a>'.format(charref), |
| 637 |
expected, collector=collector()) |
| 638 |
# check charrefs at the beginning/end of the text |
| 639 |
expected = [('data', '"'), ('starttag', 'a', []), |
| 640 |
('data', '"'), ('endtag', 'a'), ('data', '"')] |
| 641 |
for charref in charrefs: |
| 642 |
self._run_check('{0}<a>' |
| 643 |
'{0}</a>{0}'.format(charref), |
| 644 |
expected, collector=collector()) |
| 645 |
# check charrefs in <script>/<style> elements |
| 646 |
for charref in charrefs: |
| 647 |
text = 'X'.join([charref]*3) |
| 648 |
expected = [('data', '"'), |
| 649 |
('starttag', 'script', []), ('data', text), |
| 650 |
('endtag', 'script'), ('data', '"'), |
| 651 |
('starttag', 'style', []), ('data', text), |
| 652 |
('endtag', 'style'), ('data', '"')] |
| 653 |
self._run_check('{1}<script>{0}</script>{1}' |
| 654 |
'<style>{0}</style>{1}'.format(text, charref), |
| 655 |
expected, collector=collector()) |
| 656 |
# check truncated charrefs at the end of the file |
| 657 |
html = '&quo &# &#x' |
| 658 |
for x in range(1, len(html)): |
| 659 |
self._run_check(html[:x], [('data', html[:x])], |
| 660 |
collector=collector()) |
| 661 |
# check a string with no charrefs |
| 662 |
self._run_check('no charrefs here', [('data', 'no charrefs here')], |
| 663 |
collector=collector()) |
| 664 |
|
| 665 |
def test_convert_charrefs_in_attribute_values(self): |
| 666 |
# default value for convert_charrefs is now True |
| 667 |
collector = lambda: EventCollectorCharrefs() |
| 668 |
self.assertTrue(collector().convert_charrefs) |
| 669 |
|
| 670 |
# always unescape terminated entity refs, numeric and hex char refs: |
| 671 |
# - regardless whether they are at start, middle, end of attribute |
| 672 |
# - or followed by alphanumeric, non-alphanumeric, or equals char |
| 673 |
charrefs = ['¢', '¢', '¢', '¢', '¢'] |
| 674 |
expected = [('starttag', 'a', |
| 675 |
[('x', '¢'), ('x', 'z¢'), ('x', '¢z'), |
| 676 |
('x', 'z¢z'), ('x', '¢ z'), ('x', '¢=z')]), |
| 677 |
('endtag', 'a')] |
| 678 |
for charref in charrefs: |
| 679 |
self._run_check('<a x="{0}" x="z{0}" x="{0}z" ' |
| 680 |
' x="z{0}z" x="{0} z" x="{0}=z"></a>' |
| 681 |
.format(charref), expected, collector=collector()) |
| 682 |
|
| 683 |
# only unescape unterminated entity matches if they are not followed by |
| 684 |
# an alphanumeric or an equals sign |
| 685 |
charref = '¢' |
| 686 |
expected = [('starttag', 'a', |
| 687 |
[('x', '¢'), ('x', 'z¢'), ('x', '¢z'), |
| 688 |
('x', 'z¢z'), ('x', '¢ z'), ('x', '¢=z')]), |
| 689 |
('endtag', 'a')] |
| 690 |
self._run_check('<a x="{0}" x="z{0}" x="{0}z" ' |
| 691 |
' x="z{0}z" x="{0} z" x="{0}=z"></a>' |
| 692 |
.format(charref), expected, collector=collector()) |
| 693 |
|
| 694 |
# the remaining tests were for the "tolerant" parser (which is now |
| 695 |
# the default), and check various kind of broken markup |
| 696 |
def test_tolerant_parsing(self): |
| 697 |
self._run_check('<html <html>te>>xt&a<<bc</a></html>\n' |
| 698 |
'<img src="URL><//img></html</html>', [ |
| 699 |
('starttag', 'html', [('<html', None)]), |
| 700 |
('data', 'te>>xt'), |
| 701 |
('entityref', 'a'), |
| 702 |
('data', '<'), |
| 703 |
('starttag', 'bc<', [('a', None)]), |
| 704 |
('endtag', 'html'), |
| 705 |
('data', '\n')]) |
| 706 |
|
| 707 |
def test_starttag_junk_chars(self): |
| 708 |
self._run_check("<", [('data', '<')]) |
| 709 |
self._run_check("<>", [('data', '<>')]) |
| 710 |
self._run_check("< >", [('data', '< >')]) |
| 711 |
self._run_check("< ", [('data', '< ')]) |
| 712 |
self._run_check("</>", []) |
| 713 |
self._run_check("<$>", [('data', '<$>')]) |
| 714 |
self._run_check("</$>", [('comment', '$')]) |
| 715 |
self._run_check("</", [('data', '</')]) |
| 716 |
self._run_check("</a", []) |
| 717 |
self._run_check("</ a>", [('comment', ' a')]) |
| 718 |
self._run_check("</ a", [('comment', ' a')]) |
| 719 |
self._run_check("<a<a>", [('starttag', 'a<a', [])]) |
| 720 |
self._run_check("</a<a>", [('endtag', 'a<a')]) |
| 721 |
self._run_check("<!", [('comment', '')]) |
| 722 |
self._run_check("<a", []) |
| 723 |
self._run_check("<a foo='bar'", []) |
| 724 |
self._run_check("<a foo='bar", []) |
| 725 |
self._run_check("<a foo='>'", []) |
| 726 |
self._run_check("<a foo='>", []) |
| 727 |
self._run_check("<a$>", [('starttag', 'a$', [])]) |
| 728 |
self._run_check("<a$b>", [('starttag', 'a$b', [])]) |
| 729 |
self._run_check("<a$b/>", [('startendtag', 'a$b', [])]) |
| 730 |
self._run_check("<a$b >", [('starttag', 'a$b', [])]) |
| 731 |
self._run_check("<a$b />", [('startendtag', 'a$b', [])]) |
| 732 |
self._run_check("</a$b>", [('endtag', 'a$b')]) |
| 733 |
|
| 734 |
def test_slashes_in_starttag(self): |
| 735 |
self._run_check('<a foo="var"/>', [('startendtag', 'a', [('foo', 'var')])]) |
| 736 |
html = ('<img width=902 height=250px ' |
| 737 |
'src="/sites/default/files/images/homepage/foo.jpg" ' |
| 738 |
'/*what am I doing here*/ />') |
| 739 |
expected = [( |
| 740 |
'startendtag', 'img', |
| 741 |
[('width', '902'), ('height', '250px'), |
| 742 |
('src', '/sites/default/files/images/homepage/foo.jpg'), |
| 743 |
('*what', None), ('am', None), ('i', None), |
| 744 |
('doing', None), ('here*', None)] |
| 745 |
)] |
| 746 |
self._run_check(html, expected) |
| 747 |
html = ('<a / /foo/ / /=/ / /bar/ / />' |
| 748 |
'<a / /foo/ / /=/ / /bar/ / >') |
| 749 |
expected = [ |
| 750 |
('startendtag', 'a', [('foo', None), ('=', None), ('bar', None)]), |
| 751 |
('starttag', 'a', [('foo', None), ('=', None), ('bar', None)]) |
| 752 |
] |
| 753 |
self._run_check(html, expected) |
| 754 |
#see issue #14538 |
| 755 |
html = ('<meta><meta / ><meta // ><meta / / >' |
| 756 |
'<meta/><meta /><meta //><meta//>') |
| 757 |
expected = [ |
| 758 |
('starttag', 'meta', []), ('starttag', 'meta', []), |
| 759 |
('starttag', 'meta', []), ('starttag', 'meta', []), |
| 760 |
('startendtag', 'meta', []), ('startendtag', 'meta', []), |
| 761 |
('startendtag', 'meta', []), ('startendtag', 'meta', []), |
| 762 |
] |
| 763 |
self._run_check(html, expected) |
| 764 |
|
| 765 |
def test_slashes_in_endtag(self): |
| 766 |
self._run_check('</a/>', [('endtag', 'a')]) |
| 767 |
self._run_check('</a foo="var"/>', [('endtag', 'a')]) |
| 768 |
|
| 769 |
def test_declaration_junk_chars(self): |
| 770 |
self._run_check("<!DOCTYPE foo $ >", [('decl', 'DOCTYPE foo $ ')]) |
| 771 |
|
| 772 |
def test_illegal_declarations(self): |
| 773 |
self._run_check('<!spacer type="block" height="25">', |
| 774 |
[('comment', 'spacer type="block" height="25"')]) |
| 775 |
|
| 776 |
def test_invalid_end_tags(self): |
| 777 |
# A collection of broken end tags. <br> is used as separator. |
| 778 |
# see http://www.w3.org/TR/html5/tokenization.html#end-tag-open-state |
| 779 |
# and #13993 |
| 780 |
html = ('<br></label</p><br></div end tmAd-leaderBoard><br></<h4><br>' |
| 781 |
'</li class="unit"><br></li\r\n\t\t\t\t\t\t</ul><br></><br>') |
| 782 |
expected = [('starttag', 'br', []), |
| 783 |
# < is part of the name, / is discarded, p is an attribute |
| 784 |
('endtag', 'label<'), |
| 785 |
('starttag', 'br', []), |
| 786 |
# text and attributes are discarded |
| 787 |
('endtag', 'div'), |
| 788 |
('starttag', 'br', []), |
| 789 |
# comment because the first char after </ is not a-zA-Z |
| 790 |
('comment', '<h4'), |
| 791 |
('starttag', 'br', []), |
| 792 |
# attributes are discarded |
| 793 |
('endtag', 'li'), |
| 794 |
('starttag', 'br', []), |
| 795 |
# everything till ul (included) is discarded |
| 796 |
('endtag', 'li'), |
| 797 |
('starttag', 'br', []), |
| 798 |
# </> is ignored |
| 799 |
('starttag', 'br', [])] |
| 800 |
self._run_check(html, expected) |
| 801 |
|
| 802 |
def test_broken_invalid_end_tag(self): |
| 803 |
html = '<b>This</b attr=">"> confuses the parser' |
| 804 |
expected = [('starttag', 'b', []), |
| 805 |
('data', 'This'), |
| 806 |
('endtag', 'b'), |
| 807 |
('data', ' confuses the parser')] |
| 808 |
self._run_check(html, expected) |
| 809 |
|
| 810 |
def test_correct_detection_of_start_tags(self): |
| 811 |
# see #13273 |
| 812 |
html = ('<div style="" ><b>The <a href="some_url">rain</a> ' |
| 813 |
'<br /> in <span>Spain</span></b></div>') |
| 814 |
expected = [ |
| 815 |
('starttag', 'div', [('style', '')]), |
| 816 |
('starttag', 'b', []), |
| 817 |
('data', 'The '), |
| 818 |
('starttag', 'a', [('href', 'some_url')]), |
| 819 |
('data', 'rain'), |
| 820 |
('endtag', 'a'), |
| 821 |
('data', ' '), |
| 822 |
('startendtag', 'br', []), |
| 823 |
('data', ' in '), |
| 824 |
('starttag', 'span', []), |
| 825 |
('data', 'Spain'), |
| 826 |
('endtag', 'span'), |
| 827 |
('endtag', 'b'), |
| 828 |
('endtag', 'div') |
| 829 |
] |
| 830 |
self._run_check(html, expected) |
| 831 |
|
| 832 |
html = '<div style="", foo = "bar" ><b>The <a href="some_url">rain</a>' |
| 833 |
expected = [ |
| 834 |
('starttag', 'div', [('style', ''), (',', None), ('foo', 'bar')]), |
| 835 |
('starttag', 'b', []), |
| 836 |
('data', 'The '), |
| 837 |
('starttag', 'a', [('href', 'some_url')]), |
| 838 |
('data', 'rain'), |
| 839 |
('endtag', 'a'), |
| 840 |
] |
| 841 |
self._run_check(html, expected) |
| 842 |
|
| 843 |
def test_eof_in_comments(self): |
| 844 |
data = [ |
| 845 |
('<!--', [('comment', '')]), |
| 846 |
('<!---', [('comment', '')]), |
| 847 |
('<!----', [('comment', '')]), |
| 848 |
('<!-----', [('comment', '-')]), |
| 849 |
('<!------', [('comment', '--')]), |
| 850 |
('<!----!', [('comment', '')]), |
| 851 |
('<!---!', [('comment', '-!')]), |
| 852 |
('<!---!>', [('comment', '-!>')]), |
| 853 |
('<!--foo', [('comment', 'foo')]), |
| 854 |
('<!--foo-', [('comment', 'foo')]), |
| 855 |
('<!--foo--', [('comment', 'foo')]), |
| 856 |
('<!--foo--!', [('comment', 'foo')]), |
| 857 |
('<!--<!--', [('comment', '<!')]), |
| 858 |
('<!--<!--!', [('comment', '<!')]), |
| 859 |
] |
| 860 |
for html, expected in data: |
| 861 |
self._run_check(html, expected) |
| 862 |
|
| 863 |
def test_eof_in_declarations(self): |
| 864 |
data = [ |
| 865 |
('<!', [('comment', '')]), |
| 866 |
('<!-', [('comment', '-')]), |
| 867 |
('<![', [('comment', '[')]), |
| 868 |
('<!DOCTYPE', [('decl', 'DOCTYPE')]), |
| 869 |
('<!DOCTYPE ', [('decl', 'DOCTYPE ')]), |
| 870 |
('<!DOCTYPE html', [('decl', 'DOCTYPE html')]), |
| 871 |
('<!DOCTYPE html ', [('decl', 'DOCTYPE html ')]), |
| 872 |
('<!DOCTYPE html PUBLIC', [('decl', 'DOCTYPE html PUBLIC')]), |
| 873 |
('<!DOCTYPE html PUBLIC "foo', [('decl', 'DOCTYPE html PUBLIC "foo')]), |
| 874 |
('<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "foo', |
| 875 |
[('decl', 'DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "foo')]), |
| 876 |
] |
| 877 |
for html, expected in data: |
| 878 |
self._run_check(html, expected) |
| 879 |
|
| 880 |
@support.subTests('content', ['', 'x', 'x]', 'x]]']) |
| 881 |
def test_eof_in_cdata(self, content): |
| 882 |
self._run_check('<![CDATA[' + content, |
| 883 |
[('unknown decl', 'CDATA[' + content)]) |
| 884 |
self._run_check('<![CDATA[' + content, |
| 885 |
[('comment', '[CDATA[' + content)], |
| 886 |
collector=EventCollector(autocdata=True)) |
| 887 |
self._run_check('<svg><text y="100"><![CDATA[' + content, |
| 888 |
[('starttag', 'svg', []), |
| 889 |
('starttag', 'text', [('y', '100')]), |
| 890 |
('unknown decl', 'CDATA[' + content)]) |
| 891 |
|
| 892 |
def test_bogus_comments(self): |
| 893 |
html = ('<!ELEMENT br EMPTY>' |
| 894 |
'<! not really a comment >' |
| 895 |
'<! not a comment either -->' |
| 896 |
'<! -- close enough -->' |
| 897 |
'<!><!<-- this was an empty comment>' |
| 898 |
'<!!! another bogus comment !!!>' |
| 899 |
# see #32876 |
| 900 |
'<![with square brackets]!>' |
| 901 |
'<![\nmultiline\nbogusness\n]!>' |
| 902 |
'<![more brackets]-[and a hyphen]!>' |
| 903 |
'<![cdata[should be uppercase]]>' |
| 904 |
'<![CDATA [whitespaces are not ignored]]>' |
| 905 |
'<![CDATA]]>' # required '[' after CDATA |
| 906 |
) |
| 907 |
expected = [ |
| 908 |
('comment', 'ELEMENT br EMPTY'), |
| 909 |
('comment', ' not really a comment '), |
| 910 |
('comment', ' not a comment either --'), |
| 911 |
('comment', ' -- close enough --'), |
| 912 |
('comment', ''), |
| 913 |
('comment', '<-- this was an empty comment'), |
| 914 |
('comment', '!! another bogus comment !!!'), |
| 915 |
('comment', '[with square brackets]!'), |
| 916 |
('comment', '[\nmultiline\nbogusness\n]!'), |
| 917 |
('comment', '[more brackets]-[and a hyphen]!'), |
| 918 |
('comment', '[cdata[should be uppercase]]'), |
| 919 |
('comment', '[CDATA [whitespaces are not ignored]]'), |
| 920 |
('comment', '[CDATA]]'), |
| 921 |
] |
| 922 |
self._run_check(html, expected) |
| 923 |
|
| 924 |
def test_broken_condcoms(self): |
| 925 |
# these condcoms are missing the '--' after '<!' and before the '>' |
| 926 |
# and they are considered bogus comments according to |
| 927 |
# "8.2.4.42. Markup declaration open state" |
| 928 |
html = ('<![if !(IE)]>broken condcom<![endif]>' |
| 929 |
'<![if ! IE]><link href="favicon.tiff"/><![endif]>' |
| 930 |
'<![if !IE 6]><img src="firefox.png" /><![endif]>' |
| 931 |
'<![if !ie 6]><b>foo</b><![endif]>' |
| 932 |
'<![if (!IE)|(lt IE 9)]><img src="mammoth.bmp" /><![endif]>') |
| 933 |
expected = [ |
| 934 |
('comment', '[if !(IE)]'), |
| 935 |
('data', 'broken condcom'), |
| 936 |
('comment', '[endif]'), |
| 937 |
('comment', '[if ! IE]'), |
| 938 |
('startendtag', 'link', [('href', 'favicon.tiff')]), |
| 939 |
('comment', '[endif]'), |
| 940 |
('comment', '[if !IE 6]'), |
| 941 |
('startendtag', 'img', [('src', 'firefox.png')]), |
| 942 |
('comment', '[endif]'), |
| 943 |
('comment', '[if !ie 6]'), |
| 944 |
('starttag', 'b', []), |
| 945 |
('data', 'foo'), |
| 946 |
('endtag', 'b'), |
| 947 |
('comment', '[endif]'), |
| 948 |
('comment', '[if (!IE)|(lt IE 9)]'), |
| 949 |
('startendtag', 'img', [('src', 'mammoth.bmp')]), |
| 950 |
('comment', '[endif]') |
| 951 |
] |
| 952 |
self._run_check(html, expected) |
| 953 |
|
| 954 |
@support.subTests('content', [ |
| 955 |
'just some plain text', |
| 956 |
'<!-- not a comment -->', |
| 957 |
'¬-an-entity-ref;', |
| 958 |
"<not a='start tag'>", |
| 959 |
'', |
| 960 |
'[[I have many brackets]]', |
| 961 |
'I have a > in the middle', |
| 962 |
'I have a ]] in the middle', |
| 963 |
'] ]>', |
| 964 |
']] >', |
| 965 |
('\n' |
| 966 |
' if (a < b && a > b) {\n' |
| 967 |
' printf("[<marquee>How?</marquee>]");\n' |
| 968 |
' }\n'), |
| 969 |
]) |
| 970 |
def test_cdata_section_content(self, content): |
| 971 |
# See "13.2.5.42 Markup declaration open state", |
| 972 |
# "13.2.5.69 CDATA section state", and issue bpo-32876. |
| 973 |
html = f'<svg><text y="100"><![CDATA[{content}]]></text></svg>' |
| 974 |
expected = [ |
| 975 |
('starttag', 'svg', []), |
| 976 |
('starttag', 'text', [('y', '100')]), |
| 977 |
('unknown decl', 'CDATA[' + content), |
| 978 |
('endtag', 'text'), |
| 979 |
('endtag', 'svg'), |
| 980 |
] |
| 981 |
self._run_check(html, expected) |
| 982 |
self._run_check(html, expected, collector=EventCollector(autocdata=True)) |
| 983 |
|
| 984 |
def test_cdata_section(self): |
| 985 |
# See "13.2.5.42 Markup declaration open state". |
| 986 |
html = ('<![CDATA[foo<br>bar]]>' |
| 987 |
'<svg><text y="100"><![CDATA[foo<br>bar]]></text></svg>' |
| 988 |
'<![CDATA[foo<br>bar]]>') |
| 989 |
expected = [ |
| 990 |
('comment', '[CDATA[foo<br'), |
| 991 |
('data', 'bar]]>'), |
| 992 |
('starttag', 'svg', []), |
| 993 |
('starttag', 'text', [('y', '100')]), |
| 994 |
('unknown decl', 'CDATA[foo<br>bar'), |
| 995 |
('endtag', 'text'), |
| 996 |
('endtag', 'svg'), |
| 997 |
('comment', '[CDATA[foo<br'), |
| 998 |
('data', 'bar]]>'), |
| 999 |
] |
| 1000 |
self._run_check(html, expected, collector=EventCollector(autocdata=True)) |
| 1001 |
|
| 1002 |
def test_convert_charrefs_dropped_text(self): |
| 1003 |
# #23144: make sure that all the events are triggered when |
| 1004 |
# convert_charrefs is True, even if we don't call .close() |
| 1005 |
parser = EventCollector(convert_charrefs=True) |
| 1006 |
# before the fix, bar & baz was missing |
| 1007 |
parser.feed("foo <a>link</a> bar & baz") |
| 1008 |
self.assertEqual( |
| 1009 |
parser.get_events(), |
| 1010 |
[('data', 'foo '), ('starttag', 'a', []), ('data', 'link'), |
| 1011 |
('endtag', 'a'), ('data', ' bar & baz')] |
| 1012 |
) |
| 1013 |
|
| 1014 |
@support.requires_resource('cpu') |
| 1015 |
def test_eof_no_quadratic_complexity(self): |
| 1016 |
# Each of these examples used to take about an hour. |
| 1017 |
# Now they take a fraction of a second. |
| 1018 |
def check(source): |
| 1019 |
parser = html.parser.HTMLParser() |
| 1020 |
parser.feed(source) |
| 1021 |
parser.close() |
| 1022 |
n = 120_000 |
| 1023 |
check("<a " * n) |
| 1024 |
check("<a a=" * n) |
| 1025 |
check("</a " * 14 * n) |
| 1026 |
check("</a a=" * 11 * n) |
| 1027 |
check("<!--" * 4 * n) |
| 1028 |
check("<!" * 60 * n) |
| 1029 |
check("<?" * 19 * n) |
| 1030 |
check("</$" * 15 * n) |
| 1031 |
check("<![CDATA[" * 9 * n) |
| 1032 |
check("<!doctype" * 35 * n) |
| 1033 |
|
| 1034 |
|
| 1035 |
class AttributesTestCase(TestCaseBase): |
| 1036 |
|
| 1037 |
def test_attr_syntax(self): |
| 1038 |
output = [ |
| 1039 |
("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)]) |
| 1040 |
] |
| 1041 |
self._run_check("""<a b='v' c="v" d=v e>""", output) |
| 1042 |
self._run_check("<a foo==bar>", [('starttag', 'a', [('foo', '=bar')])]) |
| 1043 |
self._run_check("<a foo =bar>", [('starttag', 'a', [('foo', 'bar')])]) |
| 1044 |
self._run_check("<a foo\t=bar>", [('starttag', 'a', [('foo', 'bar')])]) |
| 1045 |
self._run_check("<a foo\v=bar>", [('starttag', 'a', [('foo\v', 'bar')])]) |
| 1046 |
self._run_check("<a foo\xa0=bar>", [('starttag', 'a', [('foo\xa0', 'bar')])]) |
| 1047 |
self._run_check("<a foo= bar>", [('starttag', 'a', [('foo', 'bar')])]) |
| 1048 |
self._run_check("<a foo=\tbar>", [('starttag', 'a', [('foo', 'bar')])]) |
| 1049 |
self._run_check("<a foo=\vbar>", [('starttag', 'a', [('foo', '\vbar')])]) |
| 1050 |
self._run_check("<a foo=\xa0bar>", [('starttag', 'a', [('foo', '\xa0bar')])]) |
| 1051 |
|
| 1052 |
def test_attr_values(self): |
| 1053 |
self._run_check("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""", |
| 1054 |
[("starttag", "a", [("b", "xxx\n\txxx"), |
| 1055 |
("c", "yyy\t\nyyy"), |
| 1056 |
("d", "\txyz\n")])]) |
| 1057 |
self._run_check("""<a b='' c="">""", |
| 1058 |
[("starttag", "a", [("b", ""), ("c", "")])]) |
| 1059 |
self._run_check("<a b=\tx c=\ny>", |
| 1060 |
[('starttag', 'a', [('b', 'x'), ('c', 'y')])]) |
| 1061 |
self._run_check("<a b=\v c=\xa0>", |
| 1062 |
[("starttag", "a", [("b", "\v"), ("c", "\xa0")])]) |
| 1063 |
# Regression test for SF patch #669683. |
| 1064 |
self._run_check("<e a=rgb(1,2,3)>", |
| 1065 |
[("starttag", "e", [("a", "rgb(1,2,3)")])]) |
| 1066 |
# Regression test for SF bug #921657. |
| 1067 |
self._run_check( |
| 1068 |
"<a href=mailto:[email protected]>", |
| 1069 |
[("starttag", "a", [("href", "mailto:[email protected]")])]) |
| 1070 |
|
| 1071 |
def test_attr_nonascii(self): |
| 1072 |
# see issue 7311 |
| 1073 |
self._run_check( |
| 1074 |
"<img src=/foo/bar.png alt=\u4e2d\u6587>", |
| 1075 |
[("starttag", "img", [("src", "/foo/bar.png"), |
| 1076 |
("alt", "\u4e2d\u6587")])]) |
| 1077 |
self._run_check( |
| 1078 |
"<a title='\u30c6\u30b9\u30c8' href='\u30c6\u30b9\u30c8.html'>", |
| 1079 |
[("starttag", "a", [("title", "\u30c6\u30b9\u30c8"), |
| 1080 |
("href", "\u30c6\u30b9\u30c8.html")])]) |
| 1081 |
self._run_check( |
| 1082 |
'<a title="\u30c6\u30b9\u30c8" href="\u30c6\u30b9\u30c8.html">', |
| 1083 |
[("starttag", "a", [("title", "\u30c6\u30b9\u30c8"), |
| 1084 |
("href", "\u30c6\u30b9\u30c8.html")])]) |
| 1085 |
|
| 1086 |
def test_attr_entity_replacement(self): |
| 1087 |
self._run_check( |
| 1088 |
"<a b='&><"''>", |
| 1089 |
[("starttag", "a", [("b", "&><\"'")])]) |
| 1090 |
|
| 1091 |
def test_attr_funky_names(self): |
| 1092 |
self._run_check( |
| 1093 |
"<a a.b='v' c:d=v e-f=v>", |
| 1094 |
[("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")])]) |
| 1095 |
|
| 1096 |
def test_entityrefs_in_attributes(self): |
| 1097 |
self._run_check( |
| 1098 |
"<html foo='€&aa&unsupported;'>", |
| 1099 |
[("starttag", "html", [("foo", "\u20AC&aa&unsupported;")])]) |
| 1100 |
|
| 1101 |
|
| 1102 |
def test_attr_funky_names2(self): |
| 1103 |
self._run_check( |
| 1104 |
r"<a $><b $=%><c \=/>", |
| 1105 |
[("starttag", "a", [("$", None)]), |
| 1106 |
("starttag", "b", [("$", "%")]), |
| 1107 |
("starttag", "c", [("\\", "/")])]) |
| 1108 |
|
| 1109 |
def test_entities_in_attribute_value(self): |
| 1110 |
# see #1200313 |
| 1111 |
for entity in ['&', '&', '&', '&']: |
| 1112 |
self._run_check('<a href="%s">' % entity, |
| 1113 |
[("starttag", "a", [("href", "&")])]) |
| 1114 |
self._run_check("<a href='%s'>" % entity, |
| 1115 |
[("starttag", "a", [("href", "&")])]) |
| 1116 |
self._run_check("<a href=%s>" % entity, |
| 1117 |
[("starttag", "a", [("href", "&")])]) |
| 1118 |
|
| 1119 |
def test_malformed_attributes(self): |
| 1120 |
# see #13357 |
| 1121 |
html = ( |
| 1122 |
"<a href=test'style='color:red;bad1'>test - bad1</a>" |
| 1123 |
"<a href=test'+style='color:red;ba2'>test - bad2</a>" |
| 1124 |
"<a href=test' style='color:red;bad3'>test - bad3</a>" |
| 1125 |
"<a href = test' style='color:red;bad4' >test - bad4</a>" |
| 1126 |
) |
| 1127 |
expected = [ |
| 1128 |
('starttag', 'a', [('href', "test'style='color:red;bad1'")]), |
| 1129 |
('data', 'test - bad1'), |
| 1130 |
('endtag', 'a'), |
| 1131 |
('starttag', 'a', [('href', "test'+style='color:red;ba2'")]), |
| 1132 |
('data', 'test - bad2'), |
| 1133 |
('endtag', 'a'), |
| 1134 |
('starttag', 'a', [('href', "test'\xa0style='color:red;bad3'")]), |
| 1135 |
('data', 'test - bad3'), |
| 1136 |
('endtag', 'a'), |
| 1137 |
('starttag', 'a', [('href', "test'\xa0style='color:red;bad4'")]), |
| 1138 |
('data', 'test - bad4'), |
| 1139 |
('endtag', 'a'), |
| 1140 |
] |
| 1141 |
self._run_check(html, expected) |
| 1142 |
|
| 1143 |
def test_malformed_adjacent_attributes(self): |
| 1144 |
# see #12629 |
| 1145 |
self._run_check('<x><y z=""o"" /></x>', |
| 1146 |
[('starttag', 'x', []), |
| 1147 |
('startendtag', 'y', [('z', ''), ('o""', None)]), |
| 1148 |
('endtag', 'x')]) |
| 1149 |
self._run_check('<x><y z="""" /></x>', |
| 1150 |
[('starttag', 'x', []), |
| 1151 |
('startendtag', 'y', [('z', ''), ('""', None)]), |
| 1152 |
('endtag', 'x')]) |
| 1153 |
|
| 1154 |
# see #755670 for the following 3 tests |
| 1155 |
def test_adjacent_attributes(self): |
| 1156 |
self._run_check('<a width="100%"cellspacing=0>', |
| 1157 |
[("starttag", "a", |
| 1158 |
[("width", "100%"), ("cellspacing","0")])]) |
| 1159 |
|
| 1160 |
self._run_check('<a id="foo"class="bar">', |
| 1161 |
[("starttag", "a", |
| 1162 |
[("id", "foo"), ("class","bar")])]) |
| 1163 |
|
| 1164 |
def test_missing_attribute_value(self): |
| 1165 |
self._run_check('<a v=>', |
| 1166 |
[("starttag", "a", [("v", "")])]) |
| 1167 |
|
| 1168 |
def test_javascript_attribute_value(self): |
| 1169 |
self._run_check("<a href=javascript:popup('/popup/help.html')>", |
| 1170 |
[("starttag", "a", |
| 1171 |
[("href", "javascript:popup('/popup/help.html')")])]) |
| 1172 |
|
| 1173 |
def test_end_tag_in_attribute_value(self): |
| 1174 |
# see #1745761 |
| 1175 |
self._run_check("<a href='http://www.example.org/\">;'>spam</a>", |
| 1176 |
[("starttag", "a", |
| 1177 |
[("href", "http://www.example.org/\">;")]), |
| 1178 |
("data", "spam"), ("endtag", "a")]) |
| 1179 |
|
| 1180 |
def test_with_unquoted_attributes(self): |
| 1181 |
# see #12008 |
| 1182 |
html = ("<html><body bgcolor=d0ca90 text='181008'>" |
| 1183 |
"<table cellspacing=0 cellpadding=1 width=100% ><tr>" |
| 1184 |
"<td align=left><font size=-1>" |
| 1185 |
"- <a href=/rabota/><span class=en> software-and-i</span></a>" |
| 1186 |
"- <a href='/1/'><span class=en> library</span></a></table>") |
| 1187 |
expected = [ |
| 1188 |
('starttag', 'html', []), |
| 1189 |
('starttag', 'body', [('bgcolor', 'd0ca90'), ('text', '181008')]), |
| 1190 |
('starttag', 'table', |
| 1191 |
[('cellspacing', '0'), ('cellpadding', '1'), ('width', '100%')]), |
| 1192 |
('starttag', 'tr', []), |
| 1193 |
('starttag', 'td', [('align', 'left')]), |
| 1194 |
('starttag', 'font', [('size', '-1')]), |
| 1195 |
('data', '- '), ('starttag', 'a', [('href', '/rabota/')]), |
| 1196 |
('starttag', 'span', [('class', 'en')]), ('data', ' software-and-i'), |
| 1197 |
('endtag', 'span'), ('endtag', 'a'), |
| 1198 |
('data', '- '), ('starttag', 'a', [('href', '/1/')]), |
| 1199 |
('starttag', 'span', [('class', 'en')]), ('data', ' library'), |
| 1200 |
('endtag', 'span'), ('endtag', 'a'), ('endtag', 'table') |
| 1201 |
] |
| 1202 |
self._run_check(html, expected) |
| 1203 |
|
| 1204 |
def test_comma_between_attributes(self): |
| 1205 |
# see bpo 41478 |
| 1206 |
# HTMLParser preserves duplicate attributes, leaving the task of |
| 1207 |
# removing duplicate attributes to a conformant html tree builder |
| 1208 |
html = ('<div class=bar,baz=asd>' # between attrs (unquoted) |
| 1209 |
'<div class="bar",baz="asd">' # between attrs (quoted) |
| 1210 |
'<div class=bar, baz=asd,>' # after values (unquoted) |
| 1211 |
'<div class="bar", baz="asd",>' # after values (quoted) |
| 1212 |
'<div class="bar",>' # one comma values (quoted) |
| 1213 |
'<div class=,bar baz=,asd>' # before values (unquoted) |
| 1214 |
'<div class=,"bar" baz=,"asd">' # before values (quoted) |
| 1215 |
'<div ,class=bar ,baz=asd>' # before names |
| 1216 |
'<div class,="bar" baz,="asd">' # after names |
| 1217 |
) |
| 1218 |
expected = [ |
| 1219 |
('starttag', 'div', [('class', 'bar,baz=asd'),]), |
| 1220 |
('starttag', 'div', [('class', 'bar'), (',baz', 'asd')]), |
| 1221 |
('starttag', 'div', [('class', 'bar,'), ('baz', 'asd,')]), |
| 1222 |
('starttag', 'div', [('class', 'bar'), (',', None), |
| 1223 |
('baz', 'asd'), (',', None)]), |
| 1224 |
('starttag', 'div', [('class', 'bar'), (',', None)]), |
| 1225 |
('starttag', 'div', [('class', ',bar'), ('baz', ',asd')]), |
| 1226 |
('starttag', 'div', [('class', ',"bar"'), ('baz', ',"asd"')]), |
| 1227 |
('starttag', 'div', [(',class', 'bar'), (',baz', 'asd')]), |
| 1228 |
('starttag', 'div', [('class,', 'bar'), ('baz,', 'asd')]), |
| 1229 |
] |
| 1230 |
self._run_check(html, expected) |
| 1231 |
|
| 1232 |
def test_weird_chars_in_unquoted_attribute_values(self): |
| 1233 |
self._run_check('<form action=bogus|&#()value>', [ |
| 1234 |
('starttag', 'form', |
| 1235 |
[('action', 'bogus|&#()value')])]) |
| 1236 |
|
| 1237 |
|
| 1238 |
class TestInheritance(unittest.TestCase): |
| 1239 |
|
| 1240 |
@patch("_markupbase.ParserBase.__init__") |
| 1241 |
@patch("_markupbase.ParserBase.reset") |
| 1242 |
def test_base_class_methods_called(self, super_reset_method, super_init_method): |
| 1243 |
with patch('_markupbase.ParserBase') as parser_base: |
| 1244 |
EventCollector() |
| 1245 |
super_init_method.assert_called_once() |
| 1246 |
super_reset_method.assert_called_once() |
| 1247 |
|
| 1248 |
|
| 1249 |
if __name__ == "__main__": |
| 1250 |
unittest.main() |