回答編集履歴
1
追記
test
CHANGED
|
@@ -9,3 +9,145 @@
|
|
|
9
9
|
* 右側と左側どちらか先に見つかったら、探索を終了して次のとんかつに移動ですか？それとも右側だけですか？
|
|
10
10
|
|
|
11
11
|
* 文章の端ではとんかつが見つからなくなる場合がありますが、その場合の終了条件はどうしますか？
|
|
12
|
+
|
|
13
|
+
|
|
14
|
+
|
|
15
|
+
----
|
|
16
|
+
|
|
17
|
+
|
|
18
|
+
|
|
19
|
+
```Python
|
|
20
|
+
|
|
21
|
+
# テストデータ
|
|
22
|
+
|
|
23
|
+
samples = [
|
|
24
|
+
|
|
25
|
+
"とんかつ1",
|
|
26
|
+
|
|
27
|
+
"とんかつ2",
|
|
28
|
+
|
|
29
|
+
"a",
|
|
30
|
+
|
|
31
|
+
"b",
|
|
32
|
+
|
|
33
|
+
"c",
|
|
34
|
+
|
|
35
|
+
"d",
|
|
36
|
+
|
|
37
|
+
"e",
|
|
38
|
+
|
|
39
|
+
"f",
|
|
40
|
+
|
|
41
|
+
"g",
|
|
42
|
+
|
|
43
|
+
"h",
|
|
44
|
+
|
|
45
|
+
"i",
|
|
46
|
+
|
|
47
|
+
"j",
|
|
48
|
+
|
|
49
|
+
"k",
|
|
50
|
+
|
|
51
|
+
"l",
|
|
52
|
+
|
|
53
|
+
"とんかつ3",
|
|
54
|
+
|
|
55
|
+
]
|
|
56
|
+
|
|
57
|
+
|
|
58
|
+
|
|
59
|
+
# 抽出コード
|
|
60
|
+
|
|
61
|
+
result = []
|
|
62
|
+
|
|
63
|
+
kw = "とんかつ"
|
|
64
|
+
|
|
65
|
+
MAX_REPEAT = 5
|
|
66
|
+
|
|
67
|
+
MIN_REPEAT = 2
|
|
68
|
+
|
|
69
|
+
for i, sample in enumerate(samples):
|
|
70
|
+
|
|
71
|
+
if kw in sample:
|
|
72
|
+
|
|
73
|
+
result.append([sample])
|
|
74
|
+
|
|
75
|
+
|
|
76
|
+
|
|
77
|
+
offset = 1
|
|
78
|
+
|
|
79
|
+
while True:
|
|
80
|
+
|
|
81
|
+
# マイナスのindexにならないようにする
|
|
82
|
+
|
|
83
|
+
if i - offset >= 0:
|
|
84
|
+
|
|
85
|
+
tmp = samples[i-offset]
|
|
86
|
+
|
|
87
|
+
else:
|
|
88
|
+
|
|
89
|
+
tmp = ""
|
|
90
|
+
|
|
91
|
+
result[-1].append(tmp)
|
|
92
|
+
|
|
93
|
+
|
|
94
|
+
|
|
95
|
+
# kwが含まれたら探索を終了する終了条件
|
|
96
|
+
|
|
97
|
+
if (any(kw in t for t in result[-1][1:])
|
|
98
|
+
|
|
99
|
+
and (offset >= MIN_REPEAT)
|
|
100
|
+
|
|
101
|
+
):
|
|
102
|
+
|
|
103
|
+
break
|
|
104
|
+
|
|
105
|
+
|
|
106
|
+
|
|
107
|
+
# プラス側のインデックスは例外処理で対処
|
|
108
|
+
|
|
109
|
+
try:
|
|
110
|
+
|
|
111
|
+
tmp = samples[i+offset]
|
|
112
|
+
|
|
113
|
+
except IndexError:
|
|
114
|
+
|
|
115
|
+
tmp = ""
|
|
116
|
+
|
|
117
|
+
result[-1].append(tmp)
|
|
118
|
+
|
|
119
|
+
|
|
120
|
+
|
|
121
|
+
# kwが含まれたら探索を終了する終了条件
|
|
122
|
+
|
|
123
|
+
if (any(kw in t for t in result[-1][1:])
|
|
124
|
+
|
|
125
|
+
and (offset >= MIN_REPEAT)
|
|
126
|
+
|
|
127
|
+
):
|
|
128
|
+
|
|
129
|
+
break
|
|
130
|
+
|
|
131
|
+
|
|
132
|
+
|
|
133
|
+
# 終了条件
|
|
134
|
+
|
|
135
|
+
if offset == MAX_REPEAT:
|
|
136
|
+
|
|
137
|
+
break
|
|
138
|
+
|
|
139
|
+
|
|
140
|
+
|
|
141
|
+
offset += 1
|
|
142
|
+
|
|
143
|
+
|
|
144
|
+
|
|
145
|
+
result
|
|
146
|
+
|
|
147
|
+
# [['とんかつ1', '', 'とんかつ2', ''],
|
|
148
|
+
|
|
149
|
+
# ['とんかつ2', 'とんかつ1', 'a', ''],
|
|
150
|
+
|
|
151
|
+
# ['とんかつ3', 'l', '', 'k', '', 'j', '', 'i', '', 'h', '']]
|
|
152
|
+
|
|
153
|
+
```
|