-
Notifications
You must be signed in to change notification settings - Fork 4
/
main.py
executable file
·349 lines (283 loc) · 9.94 KB
/
main.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
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
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
#!/usr/bin/env python
# -*- coding: utf-8 -*-
from google.appengine.ext import webapp, db
from google.appengine.ext.webapp import util, template
from google.appengine.api import memcache, urlfetch, memcache
from google.appengine.api.labs import taskqueue
import datetime
import os
import random
import re
import sys
from ConfigParser import SafeConfigParser
import twoauth
from twilog import twilog
sys.path.insert(0, os.path.join(os.path.dirname(__file__), 'lib'))
from markovchains import MarkovChains
### 文字コード設定 ###
stdin = sys.stdin
stdout = sys.stdout
reload(sys)
sys.setdefaultencoding('utf-8')
sys.stdin = stdin
sys.stdout = stdout
######################
## 定数（ツイートの種類）
USE_FILE = 1
USE_MARKOV = 2
"""
OAuth の各種キーを読み込む
"""
parser = SafeConfigParser()
parser.readfp(open('config.ini'))
sec = 'oauth'
consumer_key = parser.get(sec, 'consumer_key')
consumer_secret = parser.get(sec, 'consumer_secret')
access_token = parser.get(sec, 'access_token')
access_token_secret = parser.get(sec, 'access_token_secret')
sec = 'twilog'
original_id = parser.get(sec, 'original_id')
sec = 'bot'
tweet_type = int(parser.get(sec, 'tweet_type'))
markov = MarkovChains()
api = twoauth.api(consumer_key,
consumer_secret,
access_token,
access_token_secret)
class DoReply(db.Model):
flg = db.BooleanProperty()
"""
sentence.txt を読み込む
"""
sentences = []
def parse_tweet(text):
reply = re.compile(u'@[\S]+')
hashtag = re.compile(u'#[\S]+')
url = re.compile(r's?https?://[-_.!~*\'()a-zA-Z0-9;/?:@&=+$,%#]+', re.I)
text = reply.sub('', text)
text = hashtag.sub('', text)
text = url.sub('', text)
text = text.replace(u'．', u'。')
text = text.replace(u'，', u'、')
text = text.replace(u'「', '')
text = text.replace(u'」', '')
text = text.replace(u'？', u'?')
text = text.replace(u'！', u'!')
return text
"""
POST された文章を元に新しい文章生成
"""
def get_sentence(sentence):
markov.analyze_sentence(sentence)
return markov.make_sentence()
"""
GET: DB を元に生成した新しい文章を返す
POST: 投げられた文章を解析して DB に突っ込む
"""
def tweet_from_db():
markov.load_db('gquery2')
taskqueue.add(url='/task/talk')
return markov.db.fetch_new_sentence()
def analyse_sentence_to_db(sentence):
taskqueue.add(url='/learn_task',
params={'sentences': sentence})
class Since(db.Model):
id = db.IntegerProperty()
class PostTweetHandler(webapp.RequestHandler):
def get(self):
tweet = get_tweet(False)
api.status_update(tweet)
def post(self):
tweet = get_tweet(False)
api.status_update(tweet)
class ReplyTweetHandler(webapp.RequestHandler):
def get_sinceid(self):
since_id = memcache.get('since_id')
if since_id is None:
since = Since.get_by_key_name('since_id')
if since is not None:
since_id = since.id
return since_id
def set_sinceid(self, since_id):
memcache.set('since_id', since_id)
Since(key_name='since_id', id=int(since_id)).put()
def get(self):
mentions = api.mentions(since_id=self.get_sinceid())
reply_start = re.compile(u'(@.+?)\s', re.I | re.U)
last_tweet = ''
reply_temp_defeated = DoReply.get_by_key_name('id')
if reply_temp_defeated is None:
DoReply(key_name='id', flg=False).put()
if mentions is not None:
for status in mentions:
screen_name = status['user']['screen_name']
tweet = get_tweet(True)
while tweet == last_tweet:
tweet = get_tweet(True)
last_tweet = tweet
tweet = "@%s %s" %(screen_name, tweet)
last_since_id = status['id']
self.set_sinceid(last_since_id)
if (not reply_temp_defeated.flg):
api.status_update(tweet, in_reply_to_status_id=last_since_id)
class SinceIdHandler(webapp.RequestHandler):
def get(self):
self.response.out.write(str(memcache.get('since_id')))
class MainHandler(webapp.RequestHandler):
def get(self):
self.response.out.write('Hello World!')
class SettingHandler(webapp.RequestHandler):
def get(self):
reply_temp_defeated = DoReply.get_by_key_name('id')
flg = False
if reply_temp_defeated is None:
DoReply(key_name='id', flg=False).put()
else:
flg = reply_temp_defeated.flg
dic = {'reply_temp_defeated': flg}
self.response.out.write(template.render('views/settings.html', dic))
def post(self):
_reply_temp_defeated = self.request.get('reply_temp_defeated')
if _reply_temp_defeated == '1':
DoReply(key_name='id', flg=False).put()
else:
DoReply(key_name='id', flg=True).put()
reply_temp_defeated = DoReply.get_by_key_name('id')
dic = {'reply_temp_defeated': reply_temp_defeated.flg}
self.response.out.write(template.render('views/settings.html', dic))
### こっからマルコフ連鎖用 ###
"""
文章を生成してキューに貯める
"""
class ApiDbSentenceTalkTask(webapp.RequestHandler):
def post(self):
markov.load_db('gquery2')
markov.db.store_new_sentence()
"""
文章を解析して DB に保存
"""
class ApiDbSentenceLearnTask(webapp.RequestHandler):
def post(self):
text = self.request.get('sentences')
markov.load_db('gquery2')
markov.db.store_sentence(text)
"""
memcache の中身を全削除
"""
class DeleteHandler(webapp.RequestHandler):
def get(self):
memcache.flush_all()
"""
Twilog から学習
"""
class LearnTweetHandler(webapp.RequestHandler):
def get(self):
if tweet_type == USE_MARKOV:
yesterday = datetime.date.today() - datetime.timedelta(days=1)
log = twilog.Twilog()
tweets = log.get_tweets(original_id, yesterday)
for tweet in tweets:
text = parse_tweet(tweet)
sentences = text.split(u'。')
for sentence in sentences:
analyse_sentence_to_db(sentence)
"""
Twilog から学習（一括）
"""
class LearnTweetAllHandler(webapp.RequestHandler):
def get(self):
self.response.out.write(template.render('views/learn.html', {}))
def post(self):
s_year = int(self.request.get('s_year'))
s_month = int(self.request.get('s_month'))
s_day = int(self.request.get('s_day'))
e_year = int(self.request.get('e_year'))
e_month = int(self.request.get('e_month'))
e_day = int(self.request.get('e_day'))
start = datetime.date(s_year, s_month, s_day)
end = datetime.date(e_year, e_month, e_day)
while True:
if start == end:
break
else:
taskqueue.add(url='/task_alllearn',
params={'year':start.year, 'month': start.month,
'day':start.day})
start = start + datetime.timedelta(days=1)
self.response.out.write(template.render('views/learn_result.html', {}))
class LearnTweetAllTask(webapp.RequestHandler):
def post(self):
year = int(self.request.get('year'))
month = int(self.request.get('month'))
day = int(self.request.get('day'))
log = twilog.Twilog()
tweets = log.get_tweets(original_id, datetime.date(year, month, day))
for tweet in tweets:
text = parse_tweet(tweet)
sentences = text.split(u'。')
for sentence in sentences:
analyse_sentence_to_db(sentence)
### ここまでマルコフ連鎖用 ###
"""
ツイート内容を決める
マルコフ連鎖を使う場合
=====
return tweet_from_db()
=====
ファイルに書かれてる文章をランダムにつぶやく
file: sentence.txt
=====
return tweet_randomly_from_text('sentence.txt')
=====
"""
def get_tweet(_reply=False):
if _reply:
if tweet_type == USE_FILE:
return tweet_randomly_from_text('sentence.txt')
elif tweet_type == USE_MARKOV:
return tweet_from_db()
else:
if tweet_type == USE_FILE:
return tweet_randomly_from_text('sentence.txt')
elif tweet_type == USE_MARKOV:
return tweet_from_db()
def tweet_randomly_from_text(text):
if sentences == []:
sentence = []
for line in open(text).read().splitlines():
if line.startswith('%'):
if sentence != []:
sentences.append('\n'.join(sentence))
sentence = []
else:
sentence.append(line)
if sentence != []:
sentences.append('\n'.join(sentence))
return random.choice(sentences)
class OnakaSuitaHandler(webapp.RequestHandler):
## TLを監視して「おなかすいた」が見つかったら反応する
tweets = api.home_timeline()
if tweets is not None:
for status in tweets:
tweet = status['text']
if (tweet.find(u'おなかすいた') > 0):
api.status_update('おれもおれも')
def main():
application = webapp.WSGIApplication(
[('/tweet', PostTweetHandler),
('/reply', ReplyTweetHandler),
('/since_id', SinceIdHandler),
('/task/talk', ApiDbSentenceTalkTask),
('/learn_task', ApiDbSentenceLearnTask),
('/delete_memcache', DeleteHandler),
('/learn', LearnTweetHandler),
('/learn_tweet_all', LearnTweetAllHandler),
('/task_alllearn', LearnTweetAllTask),
('/settings', SettingHandler),
('/', MainHandler),
('/onakasuita', OnakaSuitaHandler),
],
debug=True)
util.run_wsgi_app(application)
if __name__ == '__main__':
main()