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
|
# txt = "zzzzzzzzzabczzzzzzzzzz"
# pat = "abczzzabc"
# m = len(pat)
# n = len(txt)
# R = [[m for __ in range(m)] for _ in range(0, 26)]
# good_suffix = [0 for _ in range(0, m+1)]
import sys
def alpha_number(char):
if char == "0" or char == "1":
return int(char)
return ord(char) - 97
def reverse(string):
return string[::-1]
def compare(string, i, end):
for j in range(end):
if i+j == end or string[i+j] != string[j]:
return j
def condense(binary, offset=0, size=2):
out = ""
for i in range(offset, len(binary)-offset, size):
slice = binary[i:i+size]
if len(slice) == size:
out += chr(97 + int(slice, 2))
return out
def gusfield(string):
z = [0 for _ in string]
z[0] = len(string)
r = 0
l = 0
for i in range(1, len(string)):
if i == 1: # base case
z[1] = compare(string, i, len(string))
if z[1] > 0:
r = z[1] + 1
l = 1
elif i > r: # Case 1
z[i] = compare(string, i, len(string))
if z[i] > 0:
q = i + z[i]
r = q - 1
l = i
elif i <= r: # Case 2
if z[i-l] < r-i: # Case 2a
z[i] = z[i-l]
else: # Case 2b
q = compare(string, i, len(string))
z[i] = q
r = q
l = i
return z
def gen_jump_table(pat):
m = len(pat)
R = [[-1 for __ in range(m)] for _ in range(0, 256)]
for j in range(m):
for i in range(j+1):
R[alpha_number(pat[i])][j] = i
return R
def gen_z_suffix(pat):
return reverse(gusfield(reverse(pat)))+[0]
# print(list(pat))
# print(R)
# print(Z)
def gen_good_suffix(pat, Z):
m = len(pat)
good_suffix = [0 for _ in range(0, m + 1)]
for i in range(m):
j = m - Z[i]
good_suffix[j] = i+1
return good_suffix
# print("g", good_suffix)
def gen_matched_prefix(pat):
m = len(pat)
matched_prefix = gusfield(pat)+[0]
for i in range(m-1, -1, -1):
matched_prefix[i] = max(matched_prefix[i], matched_prefix[i+1])
return matched_prefix
def preprocess(pat):
R = gen_jump_table(pat)
Z = gen_z_suffix(pat)
good_suffix = gen_good_suffix(pat, Z)
matched_prefix = gen_matched_prefix(pat)
return R, good_suffix, matched_prefix
def boyer_moore(pat, txt):
R, good_suffix, matched_prefix = preprocess(pat)
m = len(pat)
n = len(txt)
i = m-1
j = 0
occurrences = []
galils = 0
comps = 0
galil = False
start = 0
stop = 0
while j <= n-m:
match = pat[i] == txt[j+i]
comps += 1
if match:
if galil and stop >= i > start:
galils += 1
i = max(start-1, 0)
galil = False
if i == 0:
good_suffix_shift = m - matched_prefix[1]
j += good_suffix_shift
occurrences.append(j)
i = m-1
else:
i -= 1
else:
mismatched = txt[j + i]
bad_char_shift = i - R[alpha_number(mismatched)][i]
good_suffix_shift = 1
if good_suffix[i+1] > 0:
good_suffix_shift = m - good_suffix[i+1]
start = good_suffix[i+1] - m + i + 1
stop = good_suffix[i+1]
elif good_suffix[i+1] == 0:
good_suffix_shift = m - matched_prefix[i+1]
start = 0
stop = matched_prefix[i + 1]
best_shift = max(good_suffix_shift, bad_char_shift)
j += best_shift
galil = best_shift == good_suffix_shift
i = m-1
print(comps)
return comps, occurrences
def two_to_the(n):
return 1 << n
def chunky_search(pat, txt, factor=2):
occurrences = []
comps = 0
for offset in range(two_to_the(factor-1)):
padding = format(offset, f"0{factor-1}b") if len(pat) % factor else ""
augmented_pat = f"{pat}{padding}"
c, o = boyer_moore(condense(augmented_pat, 0, factor), condense(txt, offset, factor))
comps += c
print(offset, o)
occurrences += o
base_comps, base_occur = boyer_moore(pat, txt)
print(base_occur[:20])
print(occurrences[:10])
print("*"*20)
print(f"Chunky Optimisation: {len(occurrences)} occurences in {comps} comparisons.")
print(f"Normal: {len(base_occur)} occurences in {base_comps} comparisons.")
if base_comps > 0:
print(f"{comps * 100 / base_comps:.3f}% of normal Boyer-Moore")
print(f"{comps * 100 / 642096:.3f}% of their Boyer-Moore")
return comps, occurrences
def read_args():
with open(sys.argv[1], "r") as txt_file:
txt = txt_file.read()
with open(sys.argv[2], "r") as pat_file:
pat = pat_file.read()
return txt, pat
def output_matches(occurrences):
with open("output_binary_boyermoore.txt", "w") as file:
for o in occurrences:
file.write(f"{o}\n")
def main():
factor = 2
if len(sys.argv) < 3:
print("Not enough arguments!")
else:
txt, pat = read_args()
comps, occurrences = chunky_search(pat, txt, factor)
print(comps)
output_matches(occurrences)
main()
|
