1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
|
import gzip
import os.path
from bisect import bisect_left
import pytest
from whoosh.compat import permutations
from whoosh.compat import xrange
from whoosh.automata import fsa, glob, lev, reg
from whoosh.support.levenshtein import levenshtein
from whoosh.util import random_bytes
def test_nfa():
nfa = fsa.NFA(0)
nfa.add_transition(0, "a", 1)
nfa.add_transition(0, fsa.EPSILON, 4)
nfa.add_transition(0, "b", 1)
nfa.add_transition(1, "c", 4)
nfa.add_final_state(4)
assert nfa.accept("")
assert nfa.accept("ac")
assert nfa.accept("bc")
assert not nfa.accept("c")
def test_empty_string():
nfa = fsa.NFA(1)
nfa.add_final_state(1)
assert nfa.accept("")
assert not nfa.accept("a")
dfa = nfa.to_dfa()
assert dfa.accept("")
assert not dfa.accept("a")
def test_nfa2():
nfa = fsa.NFA(1)
nfa.add_transition(1, "a", 2)
nfa.add_transition(1, "c", 4)
nfa.add_transition(2, "b", 3)
nfa.add_transition(2, fsa.EPSILON, 1)
nfa.add_transition(3, "a", 2)
nfa.add_transition(4, "c", 3)
nfa.add_transition(4, fsa.EPSILON, 3)
nfa.add_final_state(3)
assert nfa.accept("ab")
assert nfa.accept("abab")
assert nfa.accept("cc")
assert nfa.accept("c")
assert nfa.accept("ccab")
assert nfa.accept("ccacc")
assert nfa.accept("ccac")
assert nfa.accept("abacab")
assert not nfa.accept("b")
assert not nfa.accept("a")
assert not nfa.accept("cb")
assert not nfa.accept("caa")
dfa = nfa.to_dfa()
assert dfa.accept("ab")
assert dfa.accept("abab")
assert dfa.accept("cc")
assert dfa.accept("c")
assert dfa.accept("ccab")
assert dfa.accept("ccacc")
assert dfa.accept("ccac")
assert dfa.accept("abacab")
assert not dfa.accept("b")
assert not dfa.accept("a")
assert not dfa.accept("cb")
assert not dfa.accept("caa")
def test_insert():
nfa1 = fsa.NFA(1)
nfa1.add_transition(1, "a", 2)
nfa1.add_transition(2, "b", 3)
nfa1.add_final_state(3)
nfa2 = fsa.NFA(4)
nfa2.add_transition(4, "x", 5)
nfa2.add_transition(4, "y", 5)
nfa2.insert(4, nfa1, 5)
nfa2.add_final_state(5)
assert nfa2.accept("x")
assert nfa2.accept("y")
assert nfa2.accept("ab")
assert not nfa2.accept("a")
def test_to_dfa():
nfa = fsa.NFA(0)
nfa.add_transition(0, "a", 1)
nfa.add_transition(0, fsa.EPSILON, 4)
nfa.add_transition(0, "b", 1)
nfa.add_transition(1, "c", 4)
nfa.add_final_state(4)
assert nfa.accept("")
dfa = nfa.to_dfa()
assert dfa.accept("")
assert dfa.accept("ac")
assert dfa.accept("bc")
assert not dfa.accept("c")
def test_glob_star():
nfa = glob.glob_automaton("a*c")
assert not nfa.accept("a")
assert not nfa.accept("c")
assert nfa.accept("ac")
assert nfa.accept("abc")
assert nfa.accept("abcc")
assert nfa.accept("abcac")
assert nfa.accept("aaaaaaaaaac")
assert not nfa.accept("abb")
dfa = nfa.to_dfa()
assert not dfa.accept("a")
assert not dfa.accept("c")
assert dfa.accept("ac")
assert dfa.accept("abc")
assert dfa.accept("abcc")
assert dfa.accept("abcac")
assert not dfa.accept("abb")
def test_glob_question():
nfa = glob.glob_automaton("?")
assert not nfa.accept("")
assert nfa.accept("a")
assert not nfa.accept("aa")
nfa = glob.glob_automaton("a?c")
assert not nfa.accept("a")
assert not nfa.accept("ac")
assert nfa.accept("abc")
assert not nfa.accept("aba")
def test_glob_range():
nfa = glob.glob_automaton("[ab][cd]")
assert not nfa.accept("")
assert not nfa.accept("a")
assert not nfa.accept("c")
assert nfa.accept("ac")
assert nfa.accept("bc")
assert nfa.accept("ad")
assert nfa.accept("bd")
assert not nfa.accept("acc")
# def test_glob_negate_range():
# nfa = glob.glob_automaton("a[!ab]a")
# assert not nfa.accept("aaa")
# assert not nfa.accept("aba")
# assert nfa.accept("aca")
# assert not nfa.accept("bcb")
class Skipper(object):
def __init__(self, data):
self.data = data
self.i = 0
def __call__(self, w):
if self.data[self.i] == w:
return w
self.i += 1
pos = bisect_left(self.data, w, self.i)
if pos < len(self.data):
return self.data[pos]
else:
return None
def test_levenshtein():
path = os.path.join(os.path.dirname(__file__), "english-words.10.gz")
wordfile = gzip.open(path, "rb")
words = sorted(line.decode("latin1").strip().lower() for line in wordfile)
def find_brute(target, k):
for w in words:
if levenshtein(w, target, k) <= k:
yield w
def find_auto(target, k):
dfa = lev.levenshtein_automaton(target, k).to_dfa()
sk = Skipper(words)
return fsa.find_all_matches(dfa, sk)
assert set(find_brute("look", 2)) == set(find_auto("look", 2))
assert set(find_brute("bend", 1)) == set(find_auto("bend", 1))
assert set(find_brute("puck", 1)) == set(find_auto("puck", 1))
assert set(find_brute("zero", 1)) == set(find_auto("zero", 1))
def test_levenshtein_prefix():
path = os.path.join(os.path.dirname(__file__), "english-words.10.gz")
wordfile = gzip.open(path, "rb")
words = sorted(line.decode("latin1").strip().lower() for line in wordfile)
prefixlen = 1
def find_brute(target, k):
for w in words:
d = levenshtein(w, target, k)
if d <= k and w[:prefixlen] == target[:prefixlen]:
yield w
def find_auto(target, k):
dfa = lev.levenshtein_automaton(target, k, prefix=prefixlen).to_dfa()
sk = Skipper(words)
return fsa.find_all_matches(dfa, sk)
assert set(find_brute("look", 2)) == set(find_auto("look", 2))
assert set(find_brute("bend", 1)) == set(find_auto("bend", 1))
assert set(find_brute("puck", 1)) == set(find_auto("puck", 1))
assert set(find_brute("zero", 1)) == set(find_auto("zero", 1))
def test_basics():
n = fsa.epsilon_nfa()
assert n.accept("")
assert not n.accept("a")
n = fsa.basic_nfa("a")
assert not n.accept("")
assert n.accept("a")
assert not n.accept("b")
n = fsa.dot_nfa()
assert not n.accept("")
assert n.accept("a")
assert n.accept("b")
def test_concat():
n = fsa.concat_nfa(fsa.basic_nfa("a"), fsa.basic_nfa("b"))
assert not n.accept("")
assert not n.accept("a")
assert not n.accept("aa")
assert not n.accept("b")
assert not n.accept("bb")
assert not n.accept("ba")
assert not n.accept("abc")
assert n.accept("ab")
def test_choice():
n = fsa.choice_nfa(fsa.basic_nfa("a"),
fsa.choice_nfa(fsa.basic_nfa("b"),
fsa.basic_nfa("c")))
assert not n.accept("")
assert n.accept("a")
assert n.accept("b")
assert n.accept("c")
assert not n.accept("d")
assert not n.accept("aa")
assert not n.accept("ab")
assert not n.accept("abc")
def test_star():
n = fsa.star_nfa(fsa.basic_nfa("a"))
assert n.accept("")
assert n.accept("a")
assert n.accept("aaaaaa")
assert not n.accept("b")
assert not n.accept("ab")
def test_optional():
n = fsa.concat_nfa(fsa.basic_nfa("a"), fsa.optional_nfa(fsa.basic_nfa("b")))
assert n.accept("a")
assert n.accept("ab")
assert not n.accept("")
assert not n.accept("b")
assert not n.accept("ba")
assert not n.accept("bab")
def test_reverse_nfa():
n = fsa.concat_nfa(fsa.basic_nfa("a"), fsa.basic_nfa("b"))
r = fsa.reverse_nfa(n)
assert not r.accept("")
assert not r.accept("a")
assert not r.accept("aa")
assert not r.accept("b")
assert not r.accept("bb")
assert not r.accept("ab")
assert not r.accept("abc")
assert r.accept("ba")
def test_regular():
ex = fsa.star_nfa(fsa.choice_nfa(fsa.basic_nfa("a"), fsa.basic_nfa("b")))
assert ex.accept("")
assert ex.accept("a")
assert ex.accept("aaaa")
assert ex.accept("b")
assert ex.accept("bbbb")
assert ex.accept("abab")
assert ex.accept("babb")
ex = fsa.concat_nfa(
fsa.basic_nfa("a"),
fsa.concat_nfa(
fsa.optional_nfa(fsa.basic_nfa("b")),
fsa.basic_nfa("c")
)
)
assert ex.accept("ac")
assert ex.accept("abc")
assert not ex.accept("ab")
assert not ex.accept("bc")
@pytest.mark.xfail(strict=False)
def test_minimize_dfa():
# Example from www.cs.odu.edu/~toida/nerzic/390teched/regular/fa/min-fa.html
dfa = fsa.DFA(1)
dfa.add_transition(1, "a", 3)
dfa.add_transition(1, "b", 2)
dfa.add_transition(2, "a", 4)
dfa.add_transition(2, "b", 1)
dfa.add_transition(3, "a", 5)
dfa.add_transition(3, "b", 4)
dfa.add_transition(4, "a", 4)
dfa.add_transition(4, "b", 4)
dfa.add_transition(5, "a", 3)
dfa.add_transition(5, "b", 2)
dfa.add_final_state(1)
dfa.add_final_state(5)
good = fsa.DFA(1)
good.add_transition(1, "a", 3)
good.add_transition(1, "b", 2)
good.add_transition(2, "b", 1)
good.add_transition(3, "a", 1)
good.add_final_state(1)
dfa.minimize()
assert dfa == good
def test_strings_dfa():
strings = "able alfa alpha apple bar bear beat boom boot".split()
dfa = fsa.strings_dfa(strings)
output = list(dfa.generate_all())
assert output == strings
domain = "abcd"
words = set()
for i in xrange(1, len(domain) + 1):
words.update("".join(p) for p in permutations(domain[:i]))
words = sorted(words)
dfa = fsa.strings_dfa(words)
assert list(dfa.generate_all()) == words
|
