1 | #!/usr/bin/env python2
|
2 | from __future__ import print_function
|
3 |
|
4 | from _devbuild.gen.htm8_asdl import (h8_id, h8_id_t, h8_id_str, attr_name,
|
5 | attr_value_e, attr_value_str)
|
6 |
|
7 | import unittest
|
8 | import re
|
9 |
|
10 | from typing import List, Tuple, Any
|
11 |
|
12 | from data_lang import htm8
|
13 | from doctools.util import log
|
14 |
|
15 | with open('data_lang/testdata/hello.htm8') as f:
|
16 | TEST_HTML = f.read()
|
17 |
|
18 |
|
19 | class RegexTest(unittest.TestCase):
|
20 |
|
21 | def testDotAll(self):
|
22 | # type: () -> None
|
23 |
|
24 | # Note that $ matches end of line, not end of string
|
25 | p1 = re.compile(r'.')
|
26 | print(p1.match('\n'))
|
27 |
|
28 | p2 = re.compile(r'.', re.DOTALL)
|
29 | print(p2.match('\n'))
|
30 |
|
31 | #p3 = re.compile(r'[.\n]', re.VERBOSE)
|
32 | p3 = re.compile(r'[.\n]')
|
33 | print(p3.match('\n'))
|
34 |
|
35 | print('Negation')
|
36 |
|
37 | p4 = re.compile(r'[^>]')
|
38 | print(p4.match('\n'))
|
39 |
|
40 | def testAttrRe(self):
|
41 | # type: () -> None
|
42 | _ATTR_RE = htm8._ATTR_RE
|
43 | m = _ATTR_RE.match(' empty= val')
|
44 | print(m.groups())
|
45 |
|
46 |
|
47 | class FunctionsTest(unittest.TestCase):
|
48 |
|
49 | def testFindLineNum(self):
|
50 | # type: () -> None
|
51 | s = 'foo\n' * 3
|
52 | for pos in [1, 5, 10, 50]: # out of bounds
|
53 | line_num = htm8._FindLineNum(s, pos)
|
54 | print(line_num)
|
55 |
|
56 |
|
57 | def _MakeAttrLexer(t, h, expected_tag=h8_id.StartTag):
|
58 | # type: (Any, str) -> htm8.AttrLexer
|
59 |
|
60 | lx = htm8.Lexer(h)
|
61 |
|
62 | tok_id, end_pos = lx.Read()
|
63 | t.assertEqual(expected_tag, tok_id)
|
64 |
|
65 | attr_lx = htm8.AttrLexer(h)
|
66 | attr_lx.Init(lx.TagNamePos(), end_pos)
|
67 |
|
68 | return attr_lx
|
69 |
|
70 |
|
71 | class AttrLexerTest(unittest.TestCase):
|
72 |
|
73 | def testNoAttrs(self):
|
74 | # type: () -> None
|
75 |
|
76 | # TODO: h8_id.StartTag and EndTag will expose the tag_name_pos - the
|
77 | # end of the tag name
|
78 |
|
79 | h = 'x <a>'
|
80 | lx = htm8.Lexer(h)
|
81 |
|
82 | # Skip raw data
|
83 | tok_id, end_pos = lx.Read()
|
84 | self.assertEqual(h8_id.RawData, tok_id)
|
85 |
|
86 | tok_id, end_pos = lx.Read()
|
87 | self.assertEqual(h8_id.StartTag, tok_id)
|
88 |
|
89 | attr_lx = htm8.AttrLexer(h)
|
90 | attr_lx.Init(lx.TagNamePos(), end_pos)
|
91 |
|
92 | # There is no tag
|
93 | n, name_start, name_end = attr_lx.ReadName()
|
94 | self.assertEqual(n, attr_name.Done)
|
95 | self.assertEqual(-1, name_start)
|
96 | self.assertEqual(-1, name_end)
|
97 |
|
98 | try:
|
99 | result = attr_lx.ReadValue()
|
100 | except AssertionError as e:
|
101 | print(e)
|
102 | else:
|
103 | self.fail('should have failed')
|
104 |
|
105 | def testInvalid(self):
|
106 | h = '<a !>'
|
107 | attr_lx = _MakeAttrLexer(self, h)
|
108 |
|
109 | n, name_start, name_end = attr_lx.ReadName()
|
110 | self.assertEqual(n, attr_name.Invalid)
|
111 | self.assertEqual(-1, name_start)
|
112 | self.assertEqual(-1, name_end)
|
113 |
|
114 | try:
|
115 | result = attr_lx.ReadValue()
|
116 | except AssertionError as e:
|
117 | print(e)
|
118 | else:
|
119 | self.fail('should have failed')
|
120 |
|
121 | def testEmpty(self):
|
122 | h = '<img src=/>'
|
123 | attr_lx = _MakeAttrLexer(self, h, expected_tag=h8_id.StartEndTag)
|
124 |
|
125 | n, name_start, name_end = attr_lx.ReadName()
|
126 | self.assertEqual(n, attr_name.Ok)
|
127 | self.assertEqual(5, name_start)
|
128 | self.assertEqual(8, name_end)
|
129 | self.assertEqual(False, attr_lx.next_value_is_missing)
|
130 |
|
131 | self.assertEqual(True, attr_lx.AttrNameEquals('src'))
|
132 | self.assertEqual(False, attr_lx.AttrNameEquals('srcz'))
|
133 |
|
134 | v, attr_start, attr_end = attr_lx.ReadValue()
|
135 | log('v = %s', attr_value_str(v))
|
136 | self.assertEqual(attr_value_e.Empty, v)
|
137 | self.assertEqual(-1, attr_start)
|
138 | self.assertEqual(-1, attr_end)
|
139 |
|
140 | def testMissing(self):
|
141 | h = '<img SRC/>'
|
142 | attr_lx = _MakeAttrLexer(self, h, expected_tag=h8_id.StartEndTag)
|
143 |
|
144 | n, name_start, name_end = attr_lx.ReadName()
|
145 | self.assertEqual(n, attr_name.Ok)
|
146 | self.assertEqual(5, name_start)
|
147 | self.assertEqual(8, name_end)
|
148 | self.assertEqual(True, attr_lx.next_value_is_missing)
|
149 |
|
150 | self.assertEqual(True, attr_lx.AttrNameEquals('src'))
|
151 | self.assertEqual(False, attr_lx.AttrNameEquals('srcz'))
|
152 |
|
153 | v, attr_start, attr_end = attr_lx.ReadValue()
|
154 | self.assertEqual(attr_value_e.Missing, v)
|
155 | self.assertEqual(-1, attr_start)
|
156 | self.assertEqual(-1, attr_end)
|
157 |
|
158 | def testUnquoted(self):
|
159 | # CAREFUL: /> is a StartEndTag, and / is not part of unquoted value
|
160 | h = '<a x=foo/>'
|
161 | attr_lx = _MakeAttrLexer(self, h, expected_tag=h8_id.StartEndTag)
|
162 |
|
163 | n, name_start, name_end = attr_lx.ReadName()
|
164 | self.assertEqual(n, attr_name.Ok)
|
165 | self.assertEqual(3, name_start)
|
166 | self.assertEqual(4, name_end)
|
167 |
|
168 | v, attr_start, attr_end = attr_lx.ReadValue()
|
169 |
|
170 | log('v = %s', attr_value_str(v))
|
171 | log('unquoted val %r', h[attr_start:attr_end])
|
172 |
|
173 | self.assertEqual(attr_value_e.Unquoted, v)
|
174 | self.assertEqual(5, attr_start)
|
175 | self.assertEqual(8, attr_end)
|
176 |
|
177 | def testDoubleQuoted(self):
|
178 | h = '<a x="f&">'
|
179 | attr_lx = _MakeAttrLexer(self, h, expected_tag=h8_id.StartTag)
|
180 |
|
181 | n, name_start, name_end = attr_lx.ReadName()
|
182 | self.assertEqual(n, attr_name.Ok)
|
183 | self.assertEqual(3, name_start)
|
184 | self.assertEqual(4, name_end)
|
185 |
|
186 | v, attr_start, attr_end = attr_lx.ReadValue()
|
187 |
|
188 | log('v = %s', attr_value_str(v))
|
189 | log('val %r', h[attr_start:attr_end])
|
190 |
|
191 | self.assertEqual(attr_value_e.DoubleQuoted, v)
|
192 | self.assertEqual(6, attr_start)
|
193 | self.assertEqual(8, attr_end)
|
194 |
|
195 | def testSingleQuoted(self):
|
196 | h = "<a x='&f'>"
|
197 | attr_lx = _MakeAttrLexer(self, h, expected_tag=h8_id.StartTag)
|
198 |
|
199 | n, name_start, name_end = attr_lx.ReadName()
|
200 | self.assertEqual(n, attr_name.Ok)
|
201 | self.assertEqual(3, name_start)
|
202 | self.assertEqual(4, name_end)
|
203 |
|
204 | v, attr_start, attr_end = attr_lx.ReadValue()
|
205 |
|
206 | log('v = %s', attr_value_str(v))
|
207 | log('unquoted val %r', h[attr_start:attr_end])
|
208 |
|
209 | self.assertEqual(attr_value_e.SingleQuoted, v)
|
210 | self.assertEqual(6, attr_start)
|
211 | self.assertEqual(8, attr_end)
|
212 |
|
213 | def testDoubleQuoted_Bad(self):
|
214 | h = '<a x="foo>'
|
215 | attr_lx = _MakeAttrLexer(self, h, expected_tag=h8_id.StartTag)
|
216 |
|
217 | n, name_start, name_end = attr_lx.ReadName()
|
218 | self.assertEqual(n, attr_name.Ok)
|
219 | self.assertEqual(3, name_start)
|
220 | self.assertEqual(4, name_end)
|
221 |
|
222 | try:
|
223 | v, attr_start, attr_end = attr_lx.ReadValue()
|
224 | except htm8.LexError as e:
|
225 | print(e)
|
226 | else:
|
227 | self.fail('Expected LexError')
|
228 |
|
229 | def testSingleQuoted_Bad(self):
|
230 | h = "<a x='foo>"
|
231 | attr_lx = _MakeAttrLexer(self, h, expected_tag=h8_id.StartTag)
|
232 |
|
233 | n, name_start, name_end = attr_lx.ReadName()
|
234 | self.assertEqual(n, attr_name.Ok)
|
235 | self.assertEqual(3, name_start)
|
236 | self.assertEqual(4, name_end)
|
237 |
|
238 | try:
|
239 | v, attr_start, attr_end = attr_lx.ReadValue()
|
240 | except htm8.LexError as e:
|
241 | print(e)
|
242 | else:
|
243 | self.fail('Expected LexError')
|
244 |
|
245 |
|
246 | def ValidTokenList(s, no_special_tags=False):
|
247 | # type: (str, bool) -> List[Tuple[h8_id_t, int]]
|
248 | """A wrapper that can be more easily translated to C++. Doesn't use iterators."""
|
249 |
|
250 | start_pos = 0
|
251 | tokens = []
|
252 | lx = htm8.Lexer(s, no_special_tags=no_special_tags)
|
253 | while True:
|
254 | tok_id, end_pos = lx.Read()
|
255 | tokens.append((tok_id, end_pos))
|
256 | if tok_id == h8_id.EndOfStream:
|
257 | break
|
258 | if tok_id == h8_id.Invalid:
|
259 | raise htm8.LexError(s, start_pos)
|
260 | start_pos = end_pos
|
261 | return tokens
|
262 |
|
263 |
|
264 | def Lex(h, no_special_tags=False):
|
265 | # type: (str, bool) -> List[Tuple[int, int]]
|
266 | print(repr(h))
|
267 | tokens = ValidTokenList(h, no_special_tags=no_special_tags)
|
268 | start_pos = 0
|
269 | for tok_id, end_pos in tokens:
|
270 | frag = h[start_pos:end_pos]
|
271 | log('%d %s %r', end_pos, h8_id_str(tok_id), frag)
|
272 | start_pos = end_pos
|
273 | return tokens
|
274 |
|
275 |
|
276 | class LexerTest(unittest.TestCase):
|
277 |
|
278 | # IndexLinker in devtools/make_help.py
|
279 | # <pre> sections in doc/html_help.py
|
280 | # TocExtractor in devtools/cmark.py
|
281 |
|
282 | def testPstrip(self):
|
283 | # type: () -> None
|
284 | """Remove anything like this.
|
285 |
|
286 | <p><pstrip> </pstrip></p>
|
287 | """
|
288 | pass
|
289 |
|
290 | def testCommentParse(self):
|
291 | # type: () -> None
|
292 | n = len(TEST_HTML)
|
293 | tokens = Lex(TEST_HTML)
|
294 |
|
295 | def testCommentParse2(self):
|
296 | # type: () -> None
|
297 | h = '''
|
298 | hi <!-- line 1
|
299 | line 2 --><br/>'''
|
300 | tokens = Lex(h)
|
301 |
|
302 | self.assertEqual(
|
303 | [
|
304 | (h8_id.RawData, 12),
|
305 | (h8_id.Comment, 50), # <? err ?>
|
306 | (h8_id.StartEndTag, 55),
|
307 | (h8_id.EndOfStream, 55),
|
308 | ],
|
309 | tokens)
|
310 |
|
311 | def testProcessingInstruction(self):
|
312 | # type: () -> None
|
313 | # <?xml ?> header
|
314 | h = 'hi <? err ?>'
|
315 | tokens = Lex(h)
|
316 |
|
317 | self.assertEqual(
|
318 | [
|
319 | (h8_id.RawData, 3),
|
320 | (h8_id.Processing, 12), # <? err ?>
|
321 | (h8_id.EndOfStream, 12),
|
322 | ],
|
323 | tokens)
|
324 |
|
325 | def testScriptStyle(self):
|
326 | # type: () -> None
|
327 | h = '''
|
328 | hi <script src=""> if (x < 1 && y > 2 ) { console.log(""); }
|
329 | </script>
|
330 | '''
|
331 | tokens = Lex(h)
|
332 |
|
333 | expected = [
|
334 | (h8_id.RawData, 12),
|
335 | (h8_id.StartTag, 27), # <script>
|
336 | (h8_id.HtmlCData, 78), # JavaScript code is HTML CData
|
337 | (h8_id.EndTag, 87), # </script>
|
338 | (h8_id.RawData, 96), # \n
|
339 | (h8_id.EndOfStream, 96), # \n
|
340 | ]
|
341 | self.assertEqual(expected, tokens)
|
342 |
|
343 | # Test case matching
|
344 | tokens = Lex(h.replace('script', 'scrIPT'))
|
345 | self.assertEqual(expected, tokens)
|
346 |
|
347 | def testScriptStyleXml(self):
|
348 | # type: () -> None
|
349 | h = 'hi <script src=""> < </script>'
|
350 | # XML mode
|
351 | tokens = Lex(h, no_special_tags=True)
|
352 |
|
353 | self.assertEqual(
|
354 | [
|
355 | (h8_id.RawData, 3),
|
356 | (h8_id.StartTag, 18), # <script>
|
357 | (h8_id.RawData, 19), # space
|
358 | (h8_id.CharEntity, 23), # </script>
|
359 | (h8_id.RawData, 24), # \n
|
360 | (h8_id.EndTag, 33), # \n
|
361 | (h8_id.EndOfStream, 33), # \n
|
362 | ],
|
363 | tokens)
|
364 |
|
365 | def testCData(self):
|
366 | # type: () -> None
|
367 |
|
368 | # from
|
369 | # /home/andy/src/languages/Python-3.11.5/Lib/test/xmltestdata/c14n-20/inC14N4.xml
|
370 | h = '<compute><![CDATA[value>"0" && value<"10" ?"valid":"error"]]></compute>'
|
371 | tokens = Lex(h)
|
372 |
|
373 | self.assertEqual([
|
374 | (h8_id.StartTag, 9),
|
375 | (h8_id.CData, 61),
|
376 | (h8_id.EndTag, 71),
|
377 | (h8_id.EndOfStream, 71),
|
378 | ], tokens)
|
379 |
|
380 | def testEntity(self):
|
381 | # type: () -> None
|
382 |
|
383 | # from
|
384 | # /home/andy/src/Python-3.12.4/Lib/test/xmltestdata/c14n-20/inC14N5.xml
|
385 | h = '&ent1;, &ent2;!'
|
386 |
|
387 | tokens = Lex(h)
|
388 |
|
389 | self.assertEqual([
|
390 | (h8_id.CharEntity, 6),
|
391 | (h8_id.RawData, 8),
|
392 | (h8_id.CharEntity, 14),
|
393 | (h8_id.RawData, 15),
|
394 | (h8_id.EndOfStream, 15),
|
395 | ], tokens)
|
396 |
|
397 | def testStartTag(self):
|
398 | # type: () -> None
|
399 |
|
400 | h = '<a>hi</a>'
|
401 | tokens = Lex(h)
|
402 |
|
403 | self.assertEqual([
|
404 | (h8_id.StartTag, 3),
|
405 | (h8_id.RawData, 5),
|
406 | (h8_id.EndTag, 9),
|
407 | (h8_id.EndOfStream, 9),
|
408 | ], tokens)
|
409 |
|
410 | # Make sure we don't consume too much
|
411 | h = '<a><source>1.7</source></a>'
|
412 |
|
413 | tokens = Lex(h)
|
414 |
|
415 | self.assertEqual([
|
416 | (h8_id.StartTag, 3),
|
417 | (h8_id.StartTag, 11),
|
418 | (h8_id.RawData, 14),
|
419 | (h8_id.EndTag, 23),
|
420 | (h8_id.EndTag, 27),
|
421 | (h8_id.EndOfStream, 27),
|
422 | ], tokens)
|
423 |
|
424 | return
|
425 |
|
426 | h = '''
|
427 | <configuration>
|
428 | <source>1.7</source>
|
429 | </configuration>'''
|
430 |
|
431 | tokens = Lex(h)
|
432 |
|
433 | self.assertEqual([
|
434 | (h8_id.RawData, 9),
|
435 | (h8_id.StartTag, 24),
|
436 | (h8_id.RawData, 9),
|
437 | (h8_id.EndOfStream, 9),
|
438 | ], tokens)
|
439 |
|
440 | def testBad(self):
|
441 | # type: () -> None
|
442 | h = '&'
|
443 | tokens = Lex(h)
|
444 |
|
445 | self.assertEqual([
|
446 | (h8_id.BadAmpersand, 1),
|
447 | (h8_id.EndOfStream, 1),
|
448 | ], tokens)
|
449 |
|
450 | h = '>'
|
451 | tokens = Lex(h)
|
452 |
|
453 | self.assertEqual([
|
454 | (h8_id.BadGreaterThan, 1),
|
455 | (h8_id.EndOfStream, 1),
|
456 | ], tokens)
|
457 |
|
458 | def testEndOfStream(self):
|
459 | # type: () -> None
|
460 |
|
461 | # NUL is end
|
462 | h = 'a\0b'
|
463 | tokens = Lex(h)
|
464 |
|
465 | self.assertEqual([
|
466 | (h8_id.RawData, 1),
|
467 | (h8_id.EndOfStream, 2),
|
468 | ], tokens)
|
469 |
|
470 |
|
471 | if __name__ == '__main__':
|
472 | unittest.main()
|