/
search.py
170 lines (131 loc) · 4.02 KB
/
search.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
# coding: utf-8
from pyparsing import infixNotation, opAssoc, Keyword, Word, alphas
import pickle
from nltk.stem.snowball import RussianStemmer
from make_iindex import make_iindex
from make_synonims import make_synonims
import os, sys
KARENINA_NAME = 'karenina.htm'
IINDEX_NAME = 'iindex.pkl'
PARAGRAPHS_NAME = 'paragraphs.pkl'
SYNONIMS_INPUT = 'synonims.txt'
SYNONIMS_OUTPUT = 'synonims.pkl'
def OR(docs1, docs2):
res = []
i = 0
j = 0
n = len(docs1)
m = len(docs2)
while i<n and j<m:
if docs1[i] < docs2[j]:
res.append(docs1[i])
i += 1
elif docs1[i] > docs2[j]:
res.append(docs2[j])
j += 1
else:
i += 1
while i<n:
res.append(docs1[i])
i += 1
while j<m:
res.append(docs2[j])
j += 1
return res
class BoolOperand(object):
def __init__(self,t):
global STEMMER, IINDEX, SYNONIMS
term = STEMMER.stem(t[0])
self.value = IINDEX.get(term, [])
for synonim in SYNONIMS.get(term, []):
synonim_term = STEMMER.stem(synonim)
self.value = OR(self.value, IINDEX.get(synonim_term,[]))
def calcop(self):
return self.value
class BoolBinOp(object):
def __init__(self,t):
self.larg, self.rarg = tuple(t[0][0::2])
def calcop(self):
return self.calc(self.larg.calcop(), self.rarg.calcop())
class BoolAnd(BoolBinOp):
def calc(self, docs1, docs2):
res = []
i = 0
j = 0
n = len(docs1)
m = len(docs2)
while i<n and j<m:
if docs1[i] == docs2[j]:
res.append(docs1[i])
i += 1
j += 1
elif docs1[i] < docs2[j]:
i += 1
else:
j += 1
return res
class BoolOr(BoolBinOp):
def calc(self, docs1, docs2):
return OR(docs1, docs2)
class BoolNot(object):
def __init__(self,t):
self.arg = t[0][1]
def calcop(self):
global ALL_DOC_IDS
docs = self.arg.calcop()
res = []
for d in ALL_DOC_IDS:
if not (d in docs):
res.append(d)
return res
def eval_query(query_str):
global WORD_CHARS
boolOperand = Word(WORD_CHARS)
boolOperand.setParseAction(BoolOperand)
boolExpr = infixNotation( boolOperand,
[
("not", 1, opAssoc.RIGHT, BoolNot),
("and", 2, opAssoc.LEFT, BoolAnd),
("or", 2, opAssoc.LEFT, BoolOr),
]
)
return boolExpr.parseString(query_str.lower())[0].calcop()
def init():
global IINDEX, WORD_CHARS, ALL_DOC_IDS, STEMMER
read_synonims()
inp = open(IINDEX_NAME, 'rb')
IINDEX = pickle.load(inp)
m = 0
for term, posting_list in IINDEX.items():
curr_m = max(posting_list)
if curr_m > m:
m = curr_m
inp.close()
ALL_DOC_IDS = [ i for i in xrange(1,m+1) ]
STEMMER = RussianStemmer()
WORD_CHARS = u'abcdefghijklmnopqrstuvwxyzабвгдеёжзийклмнопрстуфхцчшщьыъэюя'
def read_synonims():
global SYNONIMS
inp = open(SYNONIMS_OUTPUT, 'rb')
SYNONIMS = pickle.load(inp)
inp.close()
def read_paragraphs():
inp = open(PARAGRAPHS_NAME, 'rb')
paragraphs = pickle.load(inp)
inp.close()
return paragraphs
if __name__ == '__main__':
if len(sys.argv) != 2:
print "usage: ./search.py 'семья AND друг'"
exit()
query = sys.argv[1].decode('utf-8')
filenames = os.listdir('.')
if not (PARAGRAPHS_NAME in filenames and IINDEX_NAME in filenames):
make_iindex(KARENINA_NAME, IINDEX_NAME, PARAGRAPHS_NAME)
if not (SYNONIMS_OUTPUT in filenames):
make_synonims(SYNONIMS_INPUT, SYNONIMS_OUTPUT)
init()
paragraphs = read_paragraphs()
paragraph_nums = eval_query(query)
for num in paragraph_nums:
print num, paragraphs[num-1]