| 1 | #!/usr/bin/env python2
|
| 2 | from __future__ import print_function
|
| 3 |
|
| 4 | import unittest
|
| 5 |
|
| 6 | from lazylex import html # module under test log = html.log
|
| 7 |
|
| 8 | log = html.log
|
| 9 |
|
| 10 | with open('lazylex/testdata.html') as f:
|
| 11 | TEST_HTML = f.read()
|
| 12 |
|
| 13 |
|
| 14 | class RegexTest(unittest.TestCase):
|
| 15 |
|
| 16 | def testDotAll(self):
|
| 17 | import re
|
| 18 |
|
| 19 | # Note that $ matches end of line, not end of string
|
| 20 | p1 = re.compile(r'.')
|
| 21 | print(p1.match('\n'))
|
| 22 |
|
| 23 | p2 = re.compile(r'.', re.DOTALL)
|
| 24 | print(p2.match('\n'))
|
| 25 |
|
| 26 | #p3 = re.compile(r'[.\n]', re.VERBOSE)
|
| 27 | p3 = re.compile(r'[.\n]')
|
| 28 | print(p3.match('\n'))
|
| 29 |
|
| 30 | print('Negation')
|
| 31 |
|
| 32 | p4 = re.compile(r'[^>]')
|
| 33 | print(p4.match('\n'))
|
| 34 |
|
| 35 | def testAttrRe(self):
|
| 36 | _ATTR_RE = html._ATTR_RE
|
| 37 | m = _ATTR_RE.match(' empty= val')
|
| 38 | print(m.groups())
|
| 39 |
|
| 40 |
|
| 41 | class FunctionsTest(unittest.TestCase):
|
| 42 |
|
| 43 | def testFindLineNum(self):
|
| 44 | s = 'foo\n' * 3
|
| 45 | for pos in [1, 5, 10, 50]: # out of bounds
|
| 46 | line_num = html.FindLineNum(s, pos)
|
| 47 | print(line_num)
|
| 48 |
|
| 49 |
|
| 50 | def _MakeTagLexer(s):
|
| 51 | lex = html.TagLexer(s)
|
| 52 | lex.Reset(0, len(s))
|
| 53 | return lex
|
| 54 |
|
| 55 |
|
| 56 | def _PrintTokens(lex):
|
| 57 | log('')
|
| 58 | log('tag = %r', lex.TagName())
|
| 59 | for tok, start, end in lex.Tokens():
|
| 60 | log('%s %r', tok, lex.s[start:end])
|
| 61 |
|
| 62 |
|
| 63 | class TagLexerTest(unittest.TestCase):
|
| 64 |
|
| 65 | def testTagLexer(self):
|
| 66 | # Invalid!
|
| 67 | #lex = _MakeTagLexer('< >')
|
| 68 | #print(lex.Tag())
|
| 69 |
|
| 70 | lex = _MakeTagLexer('<a>')
|
| 71 | _PrintTokens(lex)
|
| 72 |
|
| 73 | lex = _MakeTagLexer('<a novalue>')
|
| 74 | _PrintTokens(lex)
|
| 75 |
|
| 76 | # Note: we could have a different HasAttr() method
|
| 77 | # <a novalue> means lex.Get('novalue') == ''
|
| 78 | # https://developer.mozilla.org/en-US/docs/Web/API/Element/hasAttribute
|
| 79 | self.assertEqual('', lex.GetAttrRaw('novalue'))
|
| 80 |
|
| 81 | lex = _MakeTagLexer('<a href="double quoted">')
|
| 82 | _PrintTokens(lex)
|
| 83 |
|
| 84 | self.assertEqual('double quoted', lex.GetAttrRaw('href'))
|
| 85 | self.assertEqual(None, lex.GetAttrRaw('oops'))
|
| 86 |
|
| 87 | lex = _MakeTagLexer('<a href=foo class="bar">')
|
| 88 | _PrintTokens(lex)
|
| 89 |
|
| 90 | lex = _MakeTagLexer('<a href=foo class="bar" />')
|
| 91 | _PrintTokens(lex)
|
| 92 |
|
| 93 | lex = _MakeTagLexer('<a href="?foo=1&bar=2" />')
|
| 94 | self.assertEqual('?foo=1&bar=2', lex.GetAttrRaw('href'))
|
| 95 |
|
| 96 | def testTagName(self):
|
| 97 | lex = _MakeTagLexer('<a href=foo class="bar" />')
|
| 98 | self.assertEqual('a', lex.TagName())
|
| 99 |
|
| 100 | def testAllAttrs(self):
|
| 101 | """
|
| 102 | [('key', 'value')] for all
|
| 103 | """
|
| 104 | # closed
|
| 105 | lex = _MakeTagLexer('<a href=foo class="bar" />')
|
| 106 | self.assertEqual([('href', 'foo'), ('class', 'bar')],
|
| 107 | lex.AllAttrsRaw())
|
| 108 |
|
| 109 | lex = _MakeTagLexer('<a href="?foo=1&bar=2" />')
|
| 110 | self.assertEqual([('href', '?foo=1&bar=2')], lex.AllAttrsRaw())
|
| 111 |
|
| 112 | def testEmptyMissingValues(self):
|
| 113 | # equivalent to <button disabled="">
|
| 114 | lex = _MakeTagLexer('<button disabled>')
|
| 115 | all_attrs = lex.AllAttrsRaw()
|
| 116 | self.assertEqual([('disabled', '')], all_attrs)
|
| 117 |
|
| 118 | slices = lex.AllAttrsRawSlice()
|
| 119 | log('slices %s', slices)
|
| 120 |
|
| 121 | lex = _MakeTagLexer('''<p double="" single='' empty= value missing>''')
|
| 122 | all_attrs = lex.AllAttrsRaw()
|
| 123 | self.assertEqual([
|
| 124 | ('double', ''),
|
| 125 | ('single', ''),
|
| 126 | ('empty', 'value'),
|
| 127 | ('missing', ''),
|
| 128 | ], all_attrs)
|
| 129 | # TODO: should have
|
| 130 | log('all %s', all_attrs)
|
| 131 |
|
| 132 | slices = lex.AllAttrsRawSlice()
|
| 133 | log('slices %s', slices)
|
| 134 |
|
| 135 | def testInvalidTag(self):
|
| 136 | try:
|
| 137 | lex = _MakeTagLexer('<a foo=bar !></a>')
|
| 138 | all_attrs = lex.AllAttrsRaw()
|
| 139 | except html.LexError as e:
|
| 140 | print(e)
|
| 141 | else:
|
| 142 | self.fail('Expected LexError')
|
| 143 |
|
| 144 |
|
| 145 | def _MakeAttrValueLexer(s):
|
| 146 | lex = html.AttrValueLexer(s)
|
| 147 | lex.Reset(0, len(s))
|
| 148 | return lex
|
| 149 |
|
| 150 |
|
| 151 | class AttrValueLexerTest(unittest.TestCase):
|
| 152 |
|
| 153 | def testGood(self):
|
| 154 | lex = _MakeAttrValueLexer('?foo=42&bar=99')
|
| 155 | n = lex.NumTokens()
|
| 156 | self.assertEqual(3, n)
|
| 157 |
|
| 158 |
|
| 159 | def Lex(h, no_special_tags=False):
|
| 160 | print(repr(h))
|
| 161 | tokens = html.ValidTokenList(h, no_special_tags=no_special_tags)
|
| 162 | start_pos = 0
|
| 163 | for tok_id, end_pos in tokens:
|
| 164 | frag = h[start_pos:end_pos]
|
| 165 | log('%d %s %r', end_pos, html.TokenName(tok_id), frag)
|
| 166 | start_pos = end_pos
|
| 167 | return tokens
|
| 168 |
|
| 169 |
|
| 170 | class LexerTest(unittest.TestCase):
|
| 171 |
|
| 172 | # IndexLinker in devtools/make_help.py
|
| 173 | # <pre> sections in doc/html_help.py
|
| 174 | # TocExtractor in devtools/cmark.py
|
| 175 |
|
| 176 | def testPstrip(self):
|
| 177 | """Remove anything like this.
|
| 178 |
|
| 179 | <p><pstrip> </pstrip></p>
|
| 180 | """
|
| 181 | pass
|
| 182 |
|
| 183 | def testCommentParse(self):
|
| 184 | n = len(TEST_HTML)
|
| 185 | tokens = Lex(TEST_HTML)
|
| 186 |
|
| 187 | def testCommentParse2(self):
|
| 188 |
|
| 189 | Tok = html.Tok
|
| 190 | h = '''
|
| 191 | hi <!-- line 1
|
| 192 | line 2 --><br/>'''
|
| 193 | tokens = Lex(h)
|
| 194 |
|
| 195 | self.assertEqual(
|
| 196 | [
|
| 197 | (Tok.RawData, 12),
|
| 198 | (Tok.Comment, 50), # <? err ?>
|
| 199 | (Tok.StartEndTag, 55),
|
| 200 | (Tok.EndOfStream, 55),
|
| 201 | ],
|
| 202 | tokens)
|
| 203 |
|
| 204 | def testProcessingInstruction(self):
|
| 205 | # <?xml ?> header
|
| 206 | Tok = html.Tok
|
| 207 | h = 'hi <? err ?>'
|
| 208 | tokens = Lex(h)
|
| 209 |
|
| 210 | self.assertEqual(
|
| 211 | [
|
| 212 | (Tok.RawData, 3),
|
| 213 | (Tok.Processing, 12), # <? err ?>
|
| 214 | (Tok.EndOfStream, 12),
|
| 215 | ],
|
| 216 | tokens)
|
| 217 |
|
| 218 | def testScriptStyle(self):
|
| 219 | Tok = html.Tok
|
| 220 | h = '''
|
| 221 | hi <script src=""> if (x < 1 && y > 2 ) { console.log(""); }
|
| 222 | </script>
|
| 223 | '''
|
| 224 | tokens = Lex(h)
|
| 225 |
|
| 226 | self.assertEqual(
|
| 227 | [
|
| 228 | (Tok.RawData, 12),
|
| 229 | (Tok.StartTag, 27), # <script>
|
| 230 | (Tok.HtmlCData, 78), # JavaScript code is HTML CData
|
| 231 | (Tok.EndTag, 87), # </script>
|
| 232 | (Tok.RawData, 96), # \n
|
| 233 | (Tok.EndOfStream, 96), # \n
|
| 234 | ],
|
| 235 | tokens)
|
| 236 |
|
| 237 | def testScriptStyleXml(self):
|
| 238 | Tok = html.Tok
|
| 239 | h = 'hi <script src=""> < </script>'
|
| 240 | # XML mode
|
| 241 | tokens = Lex(h, no_special_tags=True)
|
| 242 |
|
| 243 | self.assertEqual(
|
| 244 | [
|
| 245 | (Tok.RawData, 3),
|
| 246 | (Tok.StartTag, 18), # <script>
|
| 247 | (Tok.RawData, 19), # space
|
| 248 | (Tok.CharEntity, 23), # </script>
|
| 249 | (Tok.RawData, 24), # \n
|
| 250 | (Tok.EndTag, 33), # \n
|
| 251 | (Tok.EndOfStream, 33), # \n
|
| 252 | ],
|
| 253 | tokens)
|
| 254 |
|
| 255 | def testCData(self):
|
| 256 | Tok = html.Tok
|
| 257 |
|
| 258 | # from
|
| 259 | # /home/andy/src/languages/Python-3.11.5/Lib/test/xmltestdata/c14n-20/inC14N4.xml
|
| 260 | h = '<compute><![CDATA[value>"0" && value<"10" ?"valid":"error"]]></compute>'
|
| 261 | tokens = Lex(h)
|
| 262 |
|
| 263 | self.assertEqual([
|
| 264 | (Tok.StartTag, 9),
|
| 265 | (Tok.CData, 61),
|
| 266 | (Tok.EndTag, 71),
|
| 267 | (Tok.EndOfStream, 71),
|
| 268 | ], tokens)
|
| 269 |
|
| 270 | def testEntity(self):
|
| 271 | Tok = html.Tok
|
| 272 |
|
| 273 | # from
|
| 274 | # /home/andy/src/Python-3.12.4/Lib/test/xmltestdata/c14n-20/inC14N5.xml
|
| 275 | h = '&ent1;, &ent2;!'
|
| 276 |
|
| 277 | tokens = Lex(h)
|
| 278 |
|
| 279 | self.assertEqual([
|
| 280 | (Tok.CharEntity, 6),
|
| 281 | (Tok.RawData, 8),
|
| 282 | (Tok.CharEntity, 14),
|
| 283 | (Tok.RawData, 15),
|
| 284 | (Tok.EndOfStream, 15),
|
| 285 | ], tokens)
|
| 286 |
|
| 287 | def testStartTag(self):
|
| 288 | Tok = html.Tok
|
| 289 |
|
| 290 | h = '<a>hi</a>'
|
| 291 | tokens = Lex(h)
|
| 292 |
|
| 293 | self.assertEqual([
|
| 294 | (Tok.StartTag, 3),
|
| 295 | (Tok.RawData, 5),
|
| 296 | (Tok.EndTag, 9),
|
| 297 | (Tok.EndOfStream, 9),
|
| 298 | ], tokens)
|
| 299 |
|
| 300 | # Make sure we don't consume too much
|
| 301 | h = '<a><source>1.7</source></a>'
|
| 302 |
|
| 303 | tokens = Lex(h)
|
| 304 |
|
| 305 | self.assertEqual([
|
| 306 | (Tok.StartTag, 3),
|
| 307 | (Tok.StartTag, 11),
|
| 308 | (Tok.RawData, 14),
|
| 309 | (Tok.EndTag, 23),
|
| 310 | (Tok.EndTag, 27),
|
| 311 | (Tok.EndOfStream, 27),
|
| 312 | ], tokens)
|
| 313 |
|
| 314 | return
|
| 315 |
|
| 316 | h = '''
|
| 317 | <configuration>
|
| 318 | <source>1.7</source>
|
| 319 | </configuration>'''
|
| 320 |
|
| 321 | tokens = Lex(h)
|
| 322 |
|
| 323 | self.assertEqual([
|
| 324 | (Tok.RawData, 9),
|
| 325 | (Tok.StartTag, 24),
|
| 326 | (Tok.RawData, 9),
|
| 327 | (Tok.EndOfStream, 9),
|
| 328 | ], tokens)
|
| 329 |
|
| 330 | def testInvalid(self):
|
| 331 | Tok = html.Tok
|
| 332 |
|
| 333 | for s in INVALID_LEX:
|
| 334 | try:
|
| 335 | tokens = html.ValidTokenList(s)
|
| 336 | except html.LexError as e:
|
| 337 | print(e)
|
| 338 | else:
|
| 339 | self.fail('Expected LexError %r' % s)
|
| 340 |
|
| 341 |
|
| 342 | INVALID_LEX = [
|
| 343 | # Should be &
|
| 344 | '<a>&',
|
| 345 | '&', # not finished
|
| 346 | '&#', # not finished
|
| 347 | # Hm > is allowed?
|
| 348 | #'a > b',
|
| 349 | 'a < b',
|
| 350 | '<!-- unfinished comment',
|
| 351 | '<? unfinished processing',
|
| 352 | '</div bad=attr> <a> <b>',
|
| 353 |
|
| 354 | # not allowed, but 3 > 4 is allowed
|
| 355 | '<a> 3 < 4 </a>',
|
| 356 | ]
|
| 357 |
|
| 358 | INVALID_PARSE = [
|
| 359 | '<a></b>',
|
| 360 | '<a>', # missing closing tag
|
| 361 | '<meta></meta>', # this is a self-closing tag
|
| 362 | ]
|
| 363 |
|
| 364 | VALID_PARSE = [
|
| 365 | '<!DOCTYPE html>\n',
|
| 366 | '<!DOCTYPE>',
|
| 367 |
|
| 368 | # empty strings
|
| 369 | '<p x=""></p>',
|
| 370 | "<p x=''></p>",
|
| 371 |
|
| 372 | # allowed, but 3 < 4 is not allowed
|
| 373 | '<a> 3 > 4 </a>',
|
| 374 | # allowed, but 3 > 4 is not allowed
|
| 375 | '<p x="3 < 4"></p>',
|
| 376 | '<b><a href="foo">link</a></b>',
|
| 377 | '<meta><a></a>',
|
| 378 | # no attribute
|
| 379 | '<button disabled></button>',
|
| 380 | '<button disabled=></button>',
|
| 381 | '<button disabled= ></button>',
|
| 382 |
|
| 383 | # single quoted is pretty common
|
| 384 | "<a href='single'></a>",
|
| 385 |
|
| 386 | # Conceding to reality - I used these myself
|
| 387 | '<a href=ble.sh></a>',
|
| 388 | '<a href=foo.html></a>',
|
| 389 |
|
| 390 | # TODO: capitalization should be allowed
|
| 391 | #'<META><a></a>',
|
| 392 |
|
| 393 | # TODO: Test <svg> and <math> ?
|
| 394 | ]
|
| 395 |
|
| 396 | VALID_XML = [
|
| 397 | '<meta></meta>',
|
| 398 | ]
|
| 399 |
|
| 400 | INVALID_TAG_LEX = [
|
| 401 | # not allowed, but 3 < 4 is allowed
|
| 402 | '<p x="3 > 4"></p>',
|
| 403 | '<a foo=bar !></a>', # bad attr
|
| 404 |
|
| 405 | # should be escaped
|
| 406 | #'<a href="&"></a>',
|
| 407 | #'<a href=">"></a>',
|
| 408 | ]
|
| 409 |
|
| 410 |
|
| 411 | class ValidateTest(unittest.TestCase):
|
| 412 |
|
| 413 | def testInvalid(self):
|
| 414 | counters = html.Counters()
|
| 415 | for s in INVALID_LEX + INVALID_TAG_LEX:
|
| 416 | try:
|
| 417 | html.Validate(s, html.BALANCED_TAGS, counters)
|
| 418 | except html.LexError as e:
|
| 419 | print(e)
|
| 420 | else:
|
| 421 | self.fail('Expected LexError %r' % s)
|
| 422 |
|
| 423 | for s in INVALID_PARSE:
|
| 424 | try:
|
| 425 | html.Validate(s, html.BALANCED_TAGS, counters)
|
| 426 | except html.ParseError as e:
|
| 427 | print(e)
|
| 428 | else:
|
| 429 | self.fail('Expected ParseError')
|
| 430 |
|
| 431 | def testValid(self):
|
| 432 | counters = html.Counters()
|
| 433 | for s in VALID_PARSE:
|
| 434 | html.Validate(s, html.BALANCED_TAGS, counters)
|
| 435 | print('HTML5 %r' % s)
|
| 436 | print('HTML5 attrs %r' % counters.debug_attrs)
|
| 437 |
|
| 438 | def testValidXml(self):
|
| 439 | counters = html.Counters()
|
| 440 | for s in VALID_XML:
|
| 441 | html.Validate(s, html.BALANCED_TAGS | html.NO_SPECIAL_TAGS,
|
| 442 | counters)
|
| 443 | print('XML %r' % s)
|
| 444 | print('XML attrs %r' % counters.debug_attrs)
|
| 445 |
|
| 446 |
|
| 447 | if __name__ == '__main__':
|
| 448 | unittest.main()
|