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