-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcrawler.py
154 lines (137 loc) · 5.71 KB
/
crawler.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
import requests
import os
from bs4 import BeautifulSoup
from urllib.parse import urlencode, urljoin
import pandas as pd
import re
from tqdm import tqdm
from urllib3.exceptions import InsecureRequestWarning
import json
requests.packages.urllib3.disable_warnings(InsecureRequestWarning)
def parse_reviews(review):
card = review.find('div', {'class': 'apphub_CardTextContent'})
date_posted = card.find('div', {'class': 'date_posted'})
date_posted.decompose()
text = re.sub(r'[\n\t\xa0]', '', card.text)
return text
def get_reviews_per_page(url, headers):
html = requests.get(url, headers=headers, timeout=5, verify=False)
soup = BeautifulSoup(html.text, 'html.parser')
reviews = soup.find_all('div', {'class': 'apphub_Card'})
review_list = []
user_reviews_cursor = soup.find('input', {'name': 'userreviewscursor'})['value']
for review in reviews:
nick = review.find('div', {'class': 'apphub_CardContentAuthorName'})
title = review.find('div', {'class': 'title'}).text
hours = review.find('div', {'class': 'hours'}).text.split(' ')[0]
date_posted = review.find('div', {'class': 'date_posted'}).text
num_replys = review.find('div', {'class': 'apphub_CardCommentCount alignNews'}).text.replace(',', '')
content_text = parse_reviews(review)
cell = [nick.text, title, hours, date_posted, num_replys, content_text, user_reviews_cursor]
review_list.append(cell)
return review_list, user_reviews_cursor
def get_url(game_id, i, language, user_reviews_cursor):
if i == 2:
query_params = {
'userreviewscursor': user_reviews_cursor,
'userreviewsoffset': str(10 * (i - 1)),
'p': str(i),
'workshopitemspage': str(i),
'readytouseitemspage': str(i),
'mtxitemspage': str(i),
'itemspage': str(i),
'screenshotspage': str(i),
'videospage': str(i),
'artpage': str(i),
'allguidepage': str(i),
'webguidepage': str(i),
'integratedguidepage': str(i),
'discussionspage': str(i),
'numperpage': '10',
'browsefilter': 'mostrecent',
'browsefilter': 'mostrecent',
'l': language,
'appHubSubSection': '10',
'filterLanguage': language,
'searchText': '',
'maxInappropriateScore': '100',
'forceanon': '1'}
else:
query_params = {
'userreviewscursor': user_reviews_cursor,
'userreviewsoffset': str(10 * (i - 1)),
'p': str(i),
'workshopitemspage': str(i),
'readytouseitemspage': str(i),
'mtxitemspage': str(i),
'itemspage': str(i),
'screenshotspage': str(i),
'videospage': str(i),
'artpage': str(i),
'allguidepage': str(i),
'webguidepage': str(i),
'integratedguidepage': str(i),
'discussionspage': str(i),
'numperpage': '10',
'browsefilter': 'mostrecent',
'browsefilter': 'mostrecent',
'l': language,
'appHubSubSection': '10',
'appHubSubSection': '10',
'filterLanguage': language,
'searchText': '',
'maxInappropriateScore': '100',
'forceanon': '1'}
base_url = 'http://steamcommunity.com/app/'
url = urljoin(base_url, f'{game_id}/homecontent/') + '?' + urlencode(query_params)
return url
def crawl(url, game_id, num_comments, language, file_name, headers):
num_pages = (num_comments + 9) // 10
columns = ['nick', 'title', 'hour', 'date_posted', 'num_replys', 'content_text', 'user_reviews_cursor']
if os.path.exists(file_name):
reviews_past = pd.read_csv(file_name, encoding='utf-8')
user_reviews_cursor = reviews_past['user_reviews_cursor'].iloc[-1]
current_page = ((reviews_past.shape[0] + 9) // 10) + 1
print("find breakpoint, continue crawling from page", current_page)
else:
first_page_list, user_reviews_cursor = get_reviews_per_page(url, headers=headers)
first_page_df = pd.DataFrame(first_page_list, columns=columns)
first_page_df.to_csv(file_name, encoding='utf-8', index=False)
current_page = 2
with tqdm(total=num_pages - current_page + 1) as pbar:
for i in range(current_page, num_pages + 1):
url = get_url(game_id, i, language, user_reviews_cursor)
review_list, user_reviews_cursor = get_reviews_per_page(url, headers=headers)
df = pd.DataFrame(review_list, columns=columns)
df.to_csv(file_name, encoding='utf-8', mode='a', header=False, index=False)
pbar.set_description(f'crawling page {i}')
pbar.update(1)
current_page = i
return current_page
with open('configs.json', 'r') as f:
configs = json.load(f)
url = configs.get('url')
game_id = configs.get('game_id')
num_comments = configs.get('num_comments')
language = configs.get('language')
file_name = configs.get('file_name')
headers = configs.get('headers')
def main():
with open('configs.json', 'r') as f:
configs = json.load(f)
url = configs.get('url')
game_id = configs.get('game_id')
num_comments = configs.get('num_comments')
language = configs.get('language')
file_name = configs.get('file_name')
headers = configs.get('headers')
while True:
try:
current_page = crawl(url, game_id, num_comments, language, file_name, headers)
if current_page == (num_comments + 9) // 10:
break
except Exception as e:
print(e)
continue
if __name__ == '__main__':
main()