forked from yidao620c/python3-cookbook
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathsmallseg.py
More file actions
144 lines (138 loc) · 4.66 KB
/
smallseg.py
File metadata and controls
144 lines (138 loc) · 4.66 KB
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
# -*- coding: utf-8 -*-
import re
import os
import sys
class SEG(object):
def __init__(self):
_localDir=os.path.dirname(__file__)
_curpath=os.path.normpath(os.path.join(os.getcwd(),_localDir))
curpath=_curpath
self.d = {}
print >> sys.stderr,"loading dict..."
self.set([x.rstrip() for x in file(os.path.join(curpath,"main.dic")) ])
self.specialwords= set([x.rstrip().decode('utf-8') for x in file(os.path.join(curpath,"suffix.dic"))])
print >> sys.stderr,'dict ok.'
#set dictionary(a list)
def set(self,keywords):
p = self.d
q = {}
k = ''
for word in keywords:
word = (chr(11)+word).decode('utf-8')
if len(word)>5:
continue
p = self.d
ln = len(word)
for i in xrange(ln-1,-1,-1):
char = word[i].lower()
if p=='':
q[k] = {}
p = q[k]
if not (char in p):
p[char] = ''
q = p
k = char
p = p[char]
pass
def _binary_seg(self,s):
ln = len(s)
if ln==1:
return [s]
R = []
for i in xrange(ln,1,-1):
tmp = s[i-2:i]
R.append(tmp)
return R
def _pro_unreg(self,piece):
#print piece
R = []
tmp = re.sub(u"。|，|,|！|…|!|《|》|<|>|\"|'|:|：|？|\?|、|\||“|”|‘|’|；|—|（|）|·|\(|\)| "," ",piece).split()
ln1 = len(tmp)
for i in xrange(len(tmp)-1,-1,-1):
mc = re.split(r"([0-9A-Za-z\-\+#@_\.]+)",tmp[i])
for j in xrange(len(mc)-1,-1,-1):
r = mc[j]
if re.search(r"([0-9A-Za-z\-\+#@_\.]+)",r)!=None:
R.append(r)
else:
R.extend(self._binary_seg(r))
return R
def cut(self,text):
"""
"""
text = text.decode('utf-8','ignore')
p = self.d
ln = len(text)
i = ln
j = 0
z = ln
q = 0
recognised = []
mem = None
mem2 = None
while i-j>0:
t = text[i-j-1].lower()
#print i,j,t,mem
if not (t in p):
if (mem!=None) or (mem2!=None):
if mem!=None:
i,j,z = mem
mem = None
elif mem2!=None:
delta = mem2[0]-i
if delta>=1:
if (delta<5) and (re.search(ur"[\w\u2E80-\u9FFF]",t)!=None):
pre = text[i-j]
#print pre
if not (pre in self.specialwords):
i,j,z,q = mem2
del recognised[q:]
mem2 = None
p = self.d
if((i<ln) and (i<z)):
unreg_tmp = self._pro_unreg(text[i:z])
recognised.extend(unreg_tmp)
recognised.append(text[i-j:i])
#print text[i-j:i],mem2
i = i-j
z = i
j = 0
continue
j = 0
i -= 1
p = self.d
continue
p = p[t]
j+=1
if chr(11) in p:
if j<=2:
mem = i,j,z
#print text[i-1]
if (z-i<2) and (text[i-1] in self.specialwords) and ((mem2==None) or ((mem2!=None and mem2[0]-i>1))):
#print text[i-1]
mem = None
mem2 = i,j,z,len(recognised)
p = self.d
i -= 1
j = 0
continue
#print mem
p = self.d
#print i,j,z,text[i:z]
if((i<ln) and (i<z)):
unreg_tmp = self._pro_unreg(text[i:z])
recognised.extend(unreg_tmp)
recognised.append(text[i-j:i])
i = i-j
z = i
j = 0
mem = None
mem2 = None
#print mem
if mem!=None:
i,j,z = mem
recognised.extend(self._pro_unreg(text[i:z]))
recognised.append(text[i-j:i])
else:
recognised.extend(self._pro_unreg(text[i-j:z]))
return recognised