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