質問編集履歴
1
いろいろ
title
CHANGED
|
@@ -1,1 +1,1 @@
|
|
|
1
|
-
英語の単語数カウント
|
|
1
|
+
nltkを用いた英語の単語数カウント
|
body
CHANGED
|
@@ -7,303 +7,73 @@
|
|
|
7
7
|
### 該当のソースコード
|
|
8
8
|
|
|
9
9
|
```python3.
|
|
10
|
-
# -*- coding:utf-8 -*-
|
|
10
|
+
# -*- coding: utf-8 -*-
|
|
11
|
-
#
|
|
11
|
+
# 単語の出現頻度をExcelに
|
|
12
12
|
|
|
13
|
-
|
|
14
|
-
from bs4 import BeautifulSoup
|
|
15
|
-
import urllib.request
|
|
16
|
-
import urllib.parse
|
|
17
|
-
import
|
|
13
|
+
import openpyxl
|
|
18
|
-
from time import sleep
|
|
19
14
|
import os
|
|
15
|
+
import juman
|
|
20
16
|
import codecs
|
|
17
|
+
import glob
|
|
18
|
+
from collections import Counter
|
|
21
19
|
|
|
22
|
-
import ssl
|
|
23
|
-
ssl._create_default_https_context = ssl._create_unverified_context
|
|
24
|
-
|
|
25
|
-
### "次へ"のページのurl取得 1
|
|
26
|
-
### レビューのページ数分のurl(1，2,…最後 までのurl)
|
|
27
|
-
def
|
|
20
|
+
def Juman_for_Frequency(spotname):
|
|
21
|
+
try:
|
|
28
|
-
|
|
22
|
+
word = [] # 単語
|
|
29
|
-
opener = urllib.request.build_opener()
|
|
30
|
-
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
|
|
31
|
-
html = opener.open(url)
|
|
32
|
-
soup = BeautifulSoup(html,"lxml")
|
|
33
|
-
|
|
34
|
-
spotname = soup.find("h1",{"class":"ui_header h1"}).get_text().strip() # タグ指定
|
|
35
|
-
links.append(url)
|
|
36
|
-
#次のページのurl(最後の口コミのとこまで)
|
|
37
|
-
for url2 in links:
|
|
38
|
-
html = opener.open(url2)
|
|
39
|
-
soup = BeautifulSoup(html,"lxml")
|
|
40
23
|
try:
|
|
41
|
-
|
|
24
|
+
fr = codecs.open(spotname, "rb", "sjis", "ignore") #読み込み
|
|
25
|
+
texts = fr.read().split("\n") #ファイルを1行ずつリストとしてtextに
|
|
26
|
+
fr.close()
|
|
27
|
+
texts = list(filter(lambda a: a != "", texts)) #空リスト削除
|
|
28
|
+
debug = 0
|
|
29
|
+
for text in texts:
|
|
30
|
+
debug +=1
|
|
42
|
-
|
|
31
|
+
print("open_file = " + spotname + "lines = " + str(debug))
|
|
43
|
-
except:
|
|
44
|
-
try:
|
|
45
|
-
link = soup.find("a",{"class":"nav next taLnk ui_button primary"}).get("href") # タグ指定
|
|
46
|
-
|
|
32
|
+
word = word + juman.keitaiso(text) #分かち書きしたワード
|
|
47
|
-
except:
|
|
48
|
-
|
|
33
|
+
counter = Counter(word) #同じ文字列をカウントしてcounterに
|
|
49
|
-
return links, spotname
|
|
50
34
|
|
|
51
|
-
### レビューの詳細ページ取得 2
|
|
52
|
-
|
|
35
|
+
except Exception:
|
|
53
|
-
|
|
36
|
+
import traceback
|
|
54
|
-
opener = urllib.request.build_opener()
|
|
55
|
-
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
|
|
56
|
-
|
|
37
|
+
traceback.print_exc()
|
|
57
|
-
soup = BeautifulSoup(html,"lxml")
|
|
58
38
|
|
|
59
|
-
#url取得
|
|
60
|
-
for div in soup.find_all("div",{"class":"ui_column is-9"}):
|
|
61
|
-
try:
|
|
62
|
-
links.append("https://www.tripadvisor.jp" + div.a.get("href"))
|
|
63
|
-
|
|
39
|
+
except Exception:
|
|
64
|
-
print("not url")
|
|
65
|
-
return links
|
|
66
|
-
|
|
67
|
-
### 英語 1
|
|
68
|
-
def Get_review_url_en(url):
|
|
69
|
-
links = []
|
|
70
|
-
opener = urllib.request.build_opener()
|
|
71
|
-
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
|
|
72
|
-
html = opener.open(url)
|
|
73
|
-
soup = BeautifulSoup(html,"lxml")
|
|
74
|
-
|
|
75
|
-
#url取得
|
|
76
|
-
for div in soup.find_all("div",{"class":"ui_column is-9"}):
|
|
77
|
-
try:
|
|
78
|
-
links.append("https://www.tripadvisor.com" + div.a.get("href"))
|
|
79
|
-
except:
|
|
80
|
-
print("not url")
|
|
81
|
-
return links
|
|
82
|
-
|
|
83
|
-
### 英語 2
|
|
84
|
-
def Get_nextpage_url_en(url):
|
|
85
|
-
links = []
|
|
86
|
-
opener = urllib.request.build_opener()
|
|
87
|
-
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
|
|
88
|
-
html = opener.open(url)
|
|
89
|
-
soup = BeautifulSoup(html,"lxml")
|
|
90
|
-
|
|
91
|
-
spotname = soup.find("h1",{"class":"ui_header h1"}).get_text().strip() # タグ指定
|
|
92
|
-
links.append(url)
|
|
93
|
-
#次のページのurl(最後の口コミのとこまで)
|
|
94
|
-
for url2 in links:
|
|
95
|
-
html = opener.open(url2)
|
|
96
|
-
soup = BeautifulSoup(html,"lxml")
|
|
97
|
-
try:
|
|
98
|
-
link = soup.find("a",{"class":"nav next taLnk "}).get("href") # タグ指定
|
|
99
|
-
links.append("https://www.tripadvisor.com" + link)
|
|
100
|
-
except:
|
|
101
|
-
try:
|
|
102
|
-
link = soup.find("a",{"class":"nav next taLnk ui_button primary"}).get("href") # タグ指定
|
|
103
|
-
links.append("https://www.tripadvisor.com" + link)
|
|
104
|
-
except:
|
|
105
|
-
print("Not next")
|
|
106
|
-
return links, spotname
|
|
107
|
-
|
|
108
|
-
|
|
109
|
-
|
|
110
|
-
### review取得 3
|
|
111
|
-
def Get_review_data(url):
|
|
112
|
-
try:
|
|
113
|
-
title = [] #タイトル
|
|
114
|
-
review = [] #レビュー
|
|
115
|
-
opener = urllib.request.build_opener()
|
|
116
|
-
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
|
|
117
|
-
html = opener.open(url)
|
|
118
|
-
soup = BeautifulSoup(html, "lxml")
|
|
119
|
-
|
|
120
|
-
#タイトル抽出1つ
|
|
121
|
-
try:
|
|
122
|
-
t = soup.find("h1",{"class":"title"})
|
|
123
|
-
title = t.get_text().replace("\n","").replace("\r","").replace(",", "").strip()
|
|
124
|
-
except:
|
|
125
|
-
t = soup.find("span",{"class":"noQuotes"})
|
|
126
|
-
title = t.get_text().replace("\n","").replace("\r","").replace(",", "").strip()
|
|
127
|
-
# for url in soup.find_all("span",{"class":"noQuotes"}): # 必要なタグ指定
|
|
128
|
-
|
|
129
|
-
#レビュー抽出1つ
|
|
130
|
-
try:
|
|
131
|
-
r = soup.find("span",{"class":"fullText "})
|
|
132
|
-
review = r.get_text().replace(",", "").replace("\n", "").replace("\r\n", "").replace("\r","").strip()
|
|
133
|
-
except:
|
|
134
|
-
r = soup.find("p",{"class":"partial_entry"})
|
|
135
|
-
review = r.get_text().replace(",", "").replace("\n", "").replace("\r\n", "").replace("\r","").strip()
|
|
136
|
-
except:
|
|
137
40
|
import traceback
|
|
138
41
|
traceback.print_exc()
|
|
139
42
|
|
|
140
|
-
return
|
|
43
|
+
return counter
|
|
141
44
|
|
|
142
45
|
|
|
143
|
-
### 取得してきたもの全てをcsvファイルに保存 4
|
|
144
|
-
def Save_path(name):
|
|
145
|
-
path = os.getcwd() # このプログラムの場所
|
|
146
|
-
savdir = "\review\" # "review"を保存するディレクトリ
|
|
147
|
-
if os.path.isdir(path + savdir) == False: # "review"というフォルダが無ければ作成
|
|
148
|
-
os.mkdir(path + savdir)
|
|
149
|
-
filename = name + ".csv"
|
|
150
|
-
|
|
46
|
+
def Save_File(data, save_directory, save_filename):
|
|
47
|
+
#ファイルに書き込み
|
|
48
|
+
wb = openpyxl.Workbook()
|
|
151
|
-
|
|
49
|
+
ws = wb.active
|
|
152
50
|
|
|
153
|
-
def Write_file(filepath, data):
|
|
154
|
-
|
|
51
|
+
num = 1
|
|
155
|
-
|
|
52
|
+
for k,v in sorted(data.items(),key=lambda x:x[1],reverse=True):
|
|
53
|
+
key = []
|
|
156
|
-
|
|
54
|
+
key = k.split(",")
|
|
55
|
+
ws.cell(column=1, row=num).value=key[0] # 単語
|
|
56
|
+
ws.cell(column=2, row=num).value=key[1] # 品詞
|
|
157
|
-
|
|
57
|
+
ws.cell(column=3, row=num).value=v # 出現回数
|
|
58
|
+
num += 1
|
|
158
59
|
|
|
60
|
+
wb.save(save_directory + save_filename)
|
|
61
|
+
print("Save_Complete!")
|
|
159
62
|
|
|
160
63
|
|
|
64
|
+
if __name__ == '__main__':
|
|
65
|
+
path = os.getcwd() # このプログラムの場所
|
|
66
|
+
textrank_dir = "\textrank\" # "textrank"を保存するディレクトリ
|
|
67
|
+
review_dir = "\review\" # "review"ディレクトリ
|
|
68
|
+
if os.path.isdir(path + textrank_dir) == False: # "textrank"というフォルダが無ければ作成
|
|
161
|
-
|
|
69
|
+
os.mkdir(path + textrank_dir)
|
|
162
|
-
|
|
70
|
+
os.chdir("review/")
|
|
163
|
-
#各値 初期化
|
|
164
|
-
title = []
|
|
165
|
-
review = []
|
|
166
|
-
links = []
|
|
167
|
-
urls = []
|
|
168
71
|
|
|
169
|
-
print("get_nextpage_url: Start")
|
|
170
|
-
try:
|
|
171
|
-
links, spotname = Get_nextpage_url_jp(start_url)
|
|
172
|
-
except:
|
|
173
|
-
print("get_nextpage_url is error")
|
|
174
|
-
print("Total Next url = " + str(len(links)))
|
|
175
|
-
|
|
72
|
+
csv_files = glob.glob("*.csv")
|
|
176
|
-
|
|
73
|
+
save_directory = path + textrank_dir
|
|
177
|
-
for
|
|
74
|
+
for spotname in csv_files:
|
|
178
|
-
|
|
75
|
+
print(spotname)
|
|
179
|
-
|
|
76
|
+
get_data = Juman_for_Frequency(spotname)
|
|
180
|
-
except:
|
|
181
|
-
|
|
77
|
+
Save_File(get_data, save_directory, spotname)
|
|
182
|
-
print("Total review = " + str(len(urls)))
|
|
183
|
-
#urls = review数
|
|
184
|
-
print("Get_review_data: Start")
|
|
185
|
-
debug = 0
|
|
186
78
|
|
|
187
|
-
#filename指定してcsvに保存
|
|
188
|
-
try:
|
|
189
|
-
filepath = Save_path(spotname)
|
|
190
|
-
except:
|
|
191
|
-
print("Save_path is error")
|
|
192
|
-
|
|
193
|
-
#file_open
|
|
194
|
-
file_object= codecs.open(filepath, "a", "cp932", "ignore")
|
|
195
|
-
for url in urls:
|
|
196
|
-
debug += 1
|
|
197
|
-
print("review = " + str(debug))
|
|
198
|
-
|
|
199
|
-
interval = 5 + random.uniform(-3.0, 3.0)
|
|
200
|
-
print("interval 1: begin")
|
|
201
|
-
sleep(interval)
|
|
202
|
-
print("interval 1: end")
|
|
203
|
-
|
|
204
|
-
try:
|
|
205
|
-
title, review = Get_review_data(url)
|
|
206
|
-
except:
|
|
207
|
-
print("Get_review_data is error")
|
|
208
|
-
|
|
209
|
-
file_object.write("{}\s{}\n".format(str(title), str(review)))
|
|
210
|
-
|
|
211
|
-
file_object.close()
|
|
212
|
-
print("save: Complete")
|
|
213
|
-
|
|
214
|
-
### 実行部分_jp
|
|
215
|
-
def Start_extract_review_en(start_url):
|
|
216
|
-
#各値 初期化
|
|
217
|
-
title = []
|
|
218
|
-
review = []
|
|
219
|
-
links = []
|
|
220
|
-
urls = []
|
|
221
|
-
|
|
222
|
-
print("get_nextpage_url: Start")
|
|
223
|
-
try:
|
|
224
|
-
links, spotname = Get_nextpage_url_en(start_url)
|
|
225
|
-
except:
|
|
226
|
-
print("get_nextpage_url is error")
|
|
227
|
-
print("Total Next url = " + str(len(links)))
|
|
228
|
-
print("get_review_url: Start")
|
|
229
|
-
#1つずつurlを
|
|
230
|
-
for link in links:
|
|
231
|
-
try:
|
|
232
|
-
urls.extend(Get_review_url_en(link)) #必要
|
|
233
|
-
except:
|
|
234
|
-
print("get_review_url is error")
|
|
235
|
-
print("Total review = " + str(len(urls)))
|
|
236
|
-
#urls = review数
|
|
237
|
-
print("Get_review_data: Start")
|
|
238
|
-
debug = 0
|
|
239
|
-
|
|
240
|
-
#filename指定してcsvに保存
|
|
241
|
-
try:
|
|
242
|
-
filepath = Save_path(spotname)
|
|
243
|
-
except:
|
|
244
|
-
print("Save_path is error")
|
|
245
|
-
|
|
246
|
-
#file_open
|
|
247
|
-
file_object= codecs.open(filepath, "a", "cp932", "ignore")
|
|
248
|
-
for url in urls:
|
|
249
|
-
debug += 1
|
|
250
|
-
print("review = " + str(debug))
|
|
251
|
-
|
|
252
|
-
interval = 5 + random.uniform(-3.0, 3.0)
|
|
253
|
-
print("interval 1: begin")
|
|
254
|
-
sleep(interval)
|
|
255
|
-
print("interval 1: end")
|
|
256
|
-
|
|
257
|
-
try:
|
|
258
|
-
title, review = Get_review_data(url)
|
|
259
|
-
except:
|
|
260
|
-
print("Get_review_data is error")
|
|
261
|
-
|
|
262
|
-
file_object.write("{}\s{}\n".format(str(title), str(review)))
|
|
263
|
-
|
|
264
|
-
file_object.close()
|
|
265
|
-
print("save: Complete")
|
|
266
|
-
|
|
267
|
-
|
|
268
|
-
if __name__ =='__main__':
|
|
269
|
-
|
|
270
|
-
"""
|
|
271
|
-
#debug Get_next_url
|
|
272
|
-
#"次へ"のページのurl取得 1
|
|
273
|
-
#url = "https://www.tripadvisor.jp/Attraction_Review-g1022838-d1548549-Reviews-Takosenbei_no_Sato-Awaji_Awaji_shima_Hyogo_Prefecture_Kinki.html" #最初のページ
|
|
274
|
-
url = "https://www.tripadvisor.com/Attraction_Review-g1022838-d1548549-Reviews-Takosenbei_no_Sato-Awaji_Awaji_shima_Hyogo_Prefecture_Kinki.html"
|
|
275
|
-
links = Get_nextpage_url_jp(url)
|
|
276
|
-
print(links)
|
|
277
|
-
"""
|
|
278
|
-
|
|
279
|
-
"""
|
|
280
|
-
#debug Get_review_url
|
|
281
|
-
# レビューの詳細ページ取得 2
|
|
282
|
-
#url = "https://www.tripadvisor.jp/Attraction_Review-g298562-d1384635-Reviews-Suma_Rikyu_Park-Kobe_Hyogo_Prefecture_Kinki.html"
|
|
283
|
-
url = "https://www.tripadvisor.com/Attraction_Review-g1022838-d1548549-Reviews-Takosenbei_no_Sato-Awaji_Awaji_shima_Hyogo_Prefecture_Kinki.html"
|
|
284
|
-
links = Get_review_url(url)
|
|
285
|
-
print(links)
|
|
286
|
-
"""
|
|
287
|
-
|
|
288
|
-
"""
|
|
289
|
-
#debug Get_review_data
|
|
290
|
-
# review 取得 3
|
|
291
|
-
#url = "https://www.tripadvisor.jp/ShowUserReviews-g298562-d1384635-r631487360-Suma_Rikyu_Park-Kobe_Hyogo_Prefecture_Kinki.html"
|
|
292
|
-
#url = "https://www.tripadvisor.jp/ShowUserReviews-g1022838-d1548549-r631449250-Takosenbei_no_Sato-Awaji_Awaji_shima_Hyogo_Prefecture_Kinki.html"
|
|
293
|
-
url = "https://www.tripadvisor.com/ShowUserReviews-g1022838-d1548549-r536077279-Takosenbei_no_Sato-Awaji_Awaji_shima_Hyogo_Prefecture_Kinki.html"
|
|
294
|
-
title, review = Get_review_data(url)
|
|
295
|
-
print("{}\s{}".format(title,review))
|
|
296
|
-
"""
|
|
297
|
-
|
|
298
|
-
#"""
|
|
299
|
-
###手打ち###
|
|
300
|
-
url_jp = "https://www.tripadvisor.jp/Attraction_Review-g1121309-d5017432-Reviews-Amarube_Railroad_Bridge_Sorano_Eki-Kami_cho_Mikata_gun_Hyogo_Prefecture_Kinki.html"
|
|
301
|
-
############
|
|
302
|
-
|
|
303
|
-
url_en = url_jp.replace("https://www.tripadvisor.jp/", "https://www.tripadvisor.com/")
|
|
304
|
-
|
|
305
|
-
Start_extract_review_jp(url_jp)
|
|
306
|
-
Start_extract_review_en(url_en)
|
|
307
|
-
#"""
|
|
308
|
-
|
|
309
79
|
```
|