-
Notifications
You must be signed in to change notification settings - Fork 11
/
wordgen.py
executable file
·193 lines (164 loc) · 6.75 KB
/
wordgen.py
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
#!/usr/bin/env python3
import argparse
import collections
import itertools
import json
import math
import re
import secrets
import string
import shutil
N = 3
V = None
ROMAN_RE = re.compile('^(?=[MDCLXVI])M*(C[MD]|D?C*)(X[CL]|L?X*)(I[XV]|V?I*)$', re.I)
def build_ngrams():
counts = collections.defaultdict(int)
for line in open('/usr/share/dict/words'):
word = line.strip()
if not re.match(r'^[a-z]{3,}$', word):
continue
word = line.strip()
counts[' ' + word[:N-1]] += 1
for n in range(len(word) - N + 2):
counts[word[n:n+N].ljust(N)] += 1
allowed = set()
for k, v in counts.items():
if v > 7 and k.strip().isalpha():
allowed.add(k)
print(' '.join(x.replace(' ', '`') for x in sorted(allowed)))
print(len(allowed), 26 ** N + (N-1) * 26 ** 2)
assert 'fbt' not in counts, counts['fbt']
if 0:
length = 10
trials = 1 << 18
valid = 0
for _ in range(trials):
word = ''.join(secrets.choice(string.ascii_lowercase) for _ in range(length))
if ' ' + word[:N-1] in allowed and all(word[n:n+N].ljust(N) in allowed for n in range(len(word) - N + 2)):
valid += 1
print(word)
print(valid, trials)
est = int(valid/trials * 26 ** length)
print(est, math.log2(est))
if 0:
global V
V = []
valid = 0
att = 0
for word in itertools.product(string.ascii_lowercase, repeat=6):
word = ''.join(word)
att += 1
if ' ' + word[:N-1] in allowed and all(word[n:n+N].ljust(N) in allowed for n in range(len(word) - N + 2)):
valid += 1
V.append(word)
if valid % 65536 == 0:
print(word)
print('V:', valid, att)
out = [0] * 27 ** 2
assert 96 == ord('`')
for w in allowed:
w = w.replace(' ', '`').encode('ascii')
out[(w[0] - 96) * 27 + w[1] - 96] |= 1 << (w[2] - 96)
return out
class WordGenerator(object):
def __init__(self, allowed, n_words=None):
self.allowed = allowed
self.path_counts = []
self.total_paths = 0
def _prepare_path_counts(self, length):
if len(self.path_counts) != length:
path_counts = [[0] * 27 ** 2 for _ in range(length)]
suff = []
for n in range(27 ** 2):
if self.allowed[n] & 1:
path_counts[length - 1][n] = 1
suff.append('%s%s' % (chr(96 + (n // 27)), chr(96 + (n % 27))))
for level in range(length - 2, 0, -1):
for n in range(27, 27 ** 2):
count = 0
for x in range(1, 27):
if self.allowed[n] & (1 << x):
count += path_counts[level + 1][(n % 27) * 27 + x]
path_counts[level][n] = count
for n in range(27):
count = 0
for x in range(1, 27):
if self.allowed[n] & (1 << x):
count += path_counts[1][(n % 27) * 27 + x]
path_counts[0][n] = count
self.path_counts = path_counts
self.total_paths = sum(path_counts[0])
return self.path_counts
def n_bits(self, length):
self._prepare_path_counts(length)
return math.log2(self.total_paths)
def generate(self, length, chosen_path=None):
''' generate a random Englishy word '''
# pick a word at random
# or, pick a path through a DAG uniformly from all paths possible
path_counts = self._prepare_path_counts(length)
# 2) pick a path to follow
if chosen_path is None:
chosen_path = secrets.randbelow(self.total_paths)
#print('%.2f bits of entropy' % math.log(self.total_paths, 2), end=' ')
#print("chose %d/%d" % (chosen_path, self.total_paths))
if not 0 <= chosen_path < self.total_paths:
raise ValueError('chosen path %d not in [0,%d)' % (chosen_path, self.total_paths))
# 3) working forwards, pick the character that contributed our chosen_path
path = chosen_path
grams = []
for level in range(length):
if level == 0:
adj = range(len(path_counts[0])) # first level: (space, char, char)
else:
base = grams[-1] % 27 * 27
adj = (base + x for x in range(27) if self.allowed[grams[-1]] & (1 << x))
for n in adj:
#print(grams, n, path_counts[level][n], path, self.adjacency_lists[n])
if path_counts[level][n] <= path:
path -= path_counts[level][n]
else:
grams.append(n)
break
else:
print("couldn't find a successor :(", grams, level)
assert len(grams) == length, chosen_path
return ''.join(chr(c % 27 + 96) for c in grams)
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--digest', action='store_true', help='build trigrams from data file')
parser.add_argument('-a', '--all', action='store_true', help='emit all possibilities')
parser.add_argument('-l', '--line', action='store_true', help='one output per line')
parser.add_argument('length', default=9, type=int, nargs='?')
parser.add_argument('count', default=64, type=int, nargs='?')
options = parser.parse_args()
if options.digest:
allowed = build_ngrams()
with open('trigrams.json', 'w') as f:
json.dump(allowed, f)
gen = WordGenerator(json.load(open('trigrams.json')))
if options.digest and V:
for n, v in enumerate(V):
assert gen.generate(len(v), n) == v, (n, v)
cols = shutil.get_terminal_size((80, 0)).columns
if options.line:
cols = 0
if options.all:
print('Generating all {2:,} {0}-letter words with {1:.1f} bits of entropy'.format(
options.length, gen.n_bits(options.length), gen.total_paths))
else:
print('Generating {} {}-letter words with {:.1f} bits of entropy ({:,} possible)'.format(
options.count, options.length, gen.n_bits(options.length), gen.total_paths))
words = (gen.generate(options.length) for _ in range(options.count))
if options.all:
words = (gen.generate(options.length, n) for n in range(gen.total_paths))
if cols > (2 + options.length * 2):
words = list(words)
cols = 1 + (cols - options.length) // (2 + options.length)
for x in range(0, len(words), cols):
print(' '.join(words[x:x+cols]))
if len(words) % cols:
print(' '.join(words[x+cols:]))
else:
for w in words:
print(w)