回答編集履歴
1
a
test
CHANGED
|
@@ -77,3 +77,233 @@
|
|
|
77
77
|
|
|
78
78
|
|
|
79
79
|
思い通りの情報をパースするには、epub 形式とスクレイピングの勉強をしてください。
|
|
80
|
+
|
|
81
|
+
|
|
82
|
+
|
|
83
|
+
## 追記
|
|
84
|
+
|
|
85
|
+
|
|
86
|
+
|
|
87
|
+
テキストをチャプターごとに分割するサンプルコード
|
|
88
|
+
|
|
89
|
+
|
|
90
|
+
|
|
91
|
+
### ファイルを読み込む。
|
|
92
|
+
|
|
93
|
+
|
|
94
|
+
|
|
95
|
+
```python
|
|
96
|
+
|
|
97
|
+
import itertools
|
|
98
|
+
|
|
99
|
+
import os
|
|
100
|
+
|
|
101
|
+
import re
|
|
102
|
+
|
|
103
|
+
|
|
104
|
+
|
|
105
|
+
# read file.
|
|
106
|
+
|
|
107
|
+
# -----------------------------------
|
|
108
|
+
|
|
109
|
+
with open('dracula.txt') as f:
|
|
110
|
+
|
|
111
|
+
# strip empty lines.
|
|
112
|
+
|
|
113
|
+
lines = [l for l in f.read().splitlines() if l]
|
|
114
|
+
|
|
115
|
+
print('number of lines: {}'.format(len(lines))) # number of lines: 13414
|
|
116
|
+
|
|
117
|
+
```
|
|
118
|
+
|
|
119
|
+
|
|
120
|
+
|
|
121
|
+
### 解析する。
|
|
122
|
+
|
|
123
|
+
|
|
124
|
+
|
|
125
|
+
```python
|
|
126
|
+
|
|
127
|
+
chapters = []
|
|
128
|
+
|
|
129
|
+
|
|
130
|
+
|
|
131
|
+
# parse the book into chapters.
|
|
132
|
+
|
|
133
|
+
# -----------------------------------
|
|
134
|
+
|
|
135
|
+
itr = iter(lines)
|
|
136
|
+
|
|
137
|
+
|
|
138
|
+
|
|
139
|
+
# skip lines until the title 'DRACULA' is found.
|
|
140
|
+
|
|
141
|
+
for i in itertools.count(1):
|
|
142
|
+
|
|
143
|
+
line = next(itr)
|
|
144
|
+
|
|
145
|
+
if line == 'DRACULA':
|
|
146
|
+
|
|
147
|
+
break
|
|
148
|
+
|
|
149
|
+
print('{} lines skipped'.format(i))
|
|
150
|
+
|
|
151
|
+
print(line)
|
|
152
|
+
|
|
153
|
+
chapters = []
|
|
154
|
+
|
|
155
|
+
|
|
156
|
+
|
|
157
|
+
# first chapter is followed by title.
|
|
158
|
+
|
|
159
|
+
next(itr) # skip chapter no
|
|
160
|
+
|
|
161
|
+
|
|
162
|
+
|
|
163
|
+
no = 1 # chapter no
|
|
164
|
+
|
|
165
|
+
title = next(itr) # chapter title
|
|
166
|
+
|
|
167
|
+
sentences = [] # sentences of chapter
|
|
168
|
+
|
|
169
|
+
|
|
170
|
+
|
|
171
|
+
for i in itertools.count(i):
|
|
172
|
+
|
|
173
|
+
line = next(itr)
|
|
174
|
+
|
|
175
|
+
|
|
176
|
+
|
|
177
|
+
if line.find(r'THE END') != -1:
|
|
178
|
+
|
|
179
|
+
# end mark found.
|
|
180
|
+
|
|
181
|
+
chapters.append({'no': no, 'title': title, 'sentences': sentences})
|
|
182
|
+
|
|
183
|
+
break # story ends.
|
|
184
|
+
|
|
185
|
+
|
|
186
|
+
|
|
187
|
+
# check if line is CHAPTER <Roman numerals>.
|
|
188
|
+
|
|
189
|
+
if re.match(r'CHAPTER [MDCLXVI]{1,2}', line):
|
|
190
|
+
|
|
191
|
+
chapters.append({'no': no, 'title': title, 'sentences': sentences})
|
|
192
|
+
|
|
193
|
+
no += 1
|
|
194
|
+
|
|
195
|
+
title = next(itr) # chapter title is follwed by chapter number.
|
|
196
|
+
|
|
197
|
+
sentences = []
|
|
198
|
+
|
|
199
|
+
continue
|
|
200
|
+
|
|
201
|
+
|
|
202
|
+
|
|
203
|
+
sentences.append(line)
|
|
204
|
+
|
|
205
|
+
|
|
206
|
+
|
|
207
|
+
print('{} lines parsed'.format(i)) # 13036 lines parsed
|
|
208
|
+
|
|
209
|
+
```
|
|
210
|
+
|
|
211
|
+
|
|
212
|
+
|
|
213
|
+
### ファイルに書き込む。
|
|
214
|
+
|
|
215
|
+
|
|
216
|
+
|
|
217
|
+
```python
|
|
218
|
+
|
|
219
|
+
# write every chapter to files.
|
|
220
|
+
|
|
221
|
+
# -------------------------------------
|
|
222
|
+
|
|
223
|
+
output_dirpath = 'chapters'
|
|
224
|
+
|
|
225
|
+
os.makedirs(output_dirpath, exist_ok=True)
|
|
226
|
+
|
|
227
|
+
|
|
228
|
+
|
|
229
|
+
for chapter in chapters:
|
|
230
|
+
|
|
231
|
+
filename = 'Dracula-Chapter-{no}_{title}.txt'.format(
|
|
232
|
+
|
|
233
|
+
no=chapter['no'], title=chapter['title'].replace(' ', '_'))
|
|
234
|
+
|
|
235
|
+
filepath = os.path.join(output_dirpath, filename)
|
|
236
|
+
|
|
237
|
+
|
|
238
|
+
|
|
239
|
+
with open(filepath, 'w') as f:
|
|
240
|
+
|
|
241
|
+
f.write("\n".join(chapter['sentences']))
|
|
242
|
+
|
|
243
|
+
```
|
|
244
|
+
|
|
245
|
+
|
|
246
|
+
|
|
247
|
+
### 出力結果
|
|
248
|
+
|
|
249
|
+
|
|
250
|
+
|
|
251
|
+
```
|
|
252
|
+
|
|
253
|
+
chapters
|
|
254
|
+
|
|
255
|
+
├── Dracula-Chapter-10__Letter,_Dr._Seward_to_Hon._Arthur_Holmwood._.txt
|
|
256
|
+
|
|
257
|
+
├── Dracula-Chapter-11__Lucy_Westenra's_Diary._.txt
|
|
258
|
+
|
|
259
|
+
├── Dracula-Chapter-12_DR._SEWARD'S_DIARY.txt
|
|
260
|
+
|
|
261
|
+
├── Dracula-Chapter-13_DR._SEWARD'S_DIARY--_continued_..txt
|
|
262
|
+
|
|
263
|
+
├── Dracula-Chapter-14_MINA_HARKER'S_JOURNAL.txt
|
|
264
|
+
|
|
265
|
+
├── Dracula-Chapter-15_DR._SEWARD'S_DIARY--_continued_..txt
|
|
266
|
+
|
|
267
|
+
├── Dracula-Chapter-16_DR._SEWARD'S_DIARY--_continued_.txt
|
|
268
|
+
|
|
269
|
+
├── Dracula-Chapter-17_DR._SEWARD'S_DIARY--_continued_.txt
|
|
270
|
+
|
|
271
|
+
├── Dracula-Chapter-18_DR._SEWARD'S_DIARY.txt
|
|
272
|
+
|
|
273
|
+
├── Dracula-Chapter-19_JONATHAN_HARKER'S_JOURNAL.txt
|
|
274
|
+
|
|
275
|
+
├── Dracula-Chapter-1_JONATHAN_HARKER'S_JOURNAL.txt
|
|
276
|
+
|
|
277
|
+
├── Dracula-Chapter-20_JONATHAN_HARKER'S_JOURNAL.txt
|
|
278
|
+
|
|
279
|
+
├── Dracula-Chapter-21_DR._SEWARD'S_DIARY.txt
|
|
280
|
+
|
|
281
|
+
├── Dracula-Chapter-22_JONATHAN_HARKER'S_JOURNAL.txt
|
|
282
|
+
|
|
283
|
+
├── Dracula-Chapter-23_DR._SEWARD'S_DIARY.txt
|
|
284
|
+
|
|
285
|
+
├── Dracula-Chapter-24_DR._SEWARD'S_PHONOGRAPH_DIARY,_SPOKEN_BY_VAN_HELSING.txt
|
|
286
|
+
|
|
287
|
+
├── Dracula-Chapter-25_DR._SEWARD'S_DIARY.txt
|
|
288
|
+
|
|
289
|
+
├── Dracula-Chapter-26_DR._SEWARD'S_DIARY.txt
|
|
290
|
+
|
|
291
|
+
├── Dracula-Chapter-27_MINA_HARKER'S_JOURNAL.txt
|
|
292
|
+
|
|
293
|
+
├── Dracula-Chapter-2_JONATHAN_HARKER'S_JOURNAL--_continued_.txt
|
|
294
|
+
|
|
295
|
+
├── Dracula-Chapter-3_JONATHAN_HARKER'S_JOURNAL--_continued_.txt
|
|
296
|
+
|
|
297
|
+
├── Dracula-Chapter-4_JONATHAN_HARKER'S_JOURNAL--_continued_.txt
|
|
298
|
+
|
|
299
|
+
├── Dracula-Chapter-5__Letter_from_Miss_Mina_Murray_to_Miss_Lucy_Westenra._.txt
|
|
300
|
+
|
|
301
|
+
├── Dracula-Chapter-6_MINA_MURRAY'S_JOURNAL.txt
|
|
302
|
+
|
|
303
|
+
├── Dracula-Chapter-7_CUTTING_FROM_"THE_DAILYGRAPH,"_8_AUGUST.txt
|
|
304
|
+
|
|
305
|
+
├── Dracula-Chapter-8_MINA_MURRAY'S_JOURNAL.txt
|
|
306
|
+
|
|
307
|
+
└── Dracula-Chapter-9__Letter,_Mina_Harker_to_Lucy_Westenra._.txt
|
|
308
|
+
|
|
309
|
+
```
|