forked from minimaxir/facebook-page-post-scraper
-
Notifications
You must be signed in to change notification settings - Fork 1
/
get_fb_comments_from_fb.py
executable file
·188 lines (146 loc) · 7.17 KB
/
get_fb_comments_from_fb.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
import urllib2
import json
import datetime
import csv
import time
from MySQL_Actions import MySQL_Codes
import settings
app_id = settings.app_id
app_secret = settings.app_secret # DO NOT SHARE WITH ANYONE!
#file_id = settings.group_or_page_id_for_comments
access_token = app_id + "|" + app_secret
db_action = MySQL_Codes()
def work_as_scheduled(sfile_id):
scrapeFacebookPageFeedComments(sfile_id, access_token)
def reset_deleted_tag():
db_action.ID_checker_already_deleted = {}
def request_until_succeed(url):
req = urllib2.Request(url)
success = False
while success is False:
try:
response = urllib2.urlopen(req)
if response.getcode() == 200:
success = True
except Exception, e:
print e
time.sleep(5)
print "Error for URL %s: %s" % (url, datetime.datetime.now())
print "Retrying."
if '400' in str(e):
return None;
return response.read()
# Needed to write tricky unicode correctly to csv
def unicode_normalize(text):
return text.translate({ 0x2018:0x27, 0x2019:0x27, 0x201C:0x22,
0x201D:0x22, 0xa0:0x20 }).encode('utf-8')
def getFacebookCommentFeedData(status_id, access_token, num_comments):
# Construct the URL string
base = "https://graph.facebook.com/v2.6"
node = "/%s/comments" % status_id
fields = "?fields=id,message,like_count,created_time,comments,from,attachment"
parameters = "&order=chronological&limit=%s&access_token=%s" % \
(num_comments, access_token)
url = base + node + fields + parameters
# retrieve data
data = request_until_succeed(url)
if data is None:
return None
else:
return json.loads(data)
def processFacebookComment(comment, status_id, parent_id = ''):
# The status is now a Python dictionary, so for top-level items,
# we can simply call the key.
# Additionally, some items may not always exist,
# so must check for existence first
comment_id = comment['id']
comment_message = '' if 'message' not in comment else \
unicode_normalize(comment['message'])
comment_author = unicode_normalize(comment['from']['name'])
comment_likes = 0 if 'like_count' not in comment else \
comment['like_count']
if 'attachment' in comment:
attach_tag = "[[%s]]" % comment['attachment']['type'].upper()
comment_message = attach_tag if comment_message is '' else \
(comment_message.decode("utf-8") + " " + \
attach_tag).encode("utf-8")
# Time needs special care since a) it's in UTC and
# b) it's not easy to use in statistical programs.
comment_published = datetime.datetime.strptime(
comment['created_time'],'%Y-%m-%dT%H:%M:%S+0000')
comment_published = comment_published + datetime.timedelta(hours=-5) # EST
comment_published = comment_published.strftime(
'%Y-%m-%d %H:%M:%S') # best time format for spreadsheet programs
# Return a tuple of all processed data
return (comment_id, status_id, parent_id, comment_message, comment_author,
comment_published, comment_likes)
def scrapeFacebookPageFeedComments(file_id, access_token):
#with open('%s_facebook_comments.csv' % file_id, 'wb') as file:
# w = csv.writer(file)
# w.writerow(["comment_id", "status_id", "parent_id", "comment_message",
# "comment_author", "comment_published", "comment_likes"])
num_processed = 0 # keep a count on how many we've processed
scrape_starttime = datetime.datetime.now()
print "Scraping %s Comments From Posts: %s\n" % \
(file_id, scrape_starttime)
#with open('%s_facebook_statuses.csv' % file_id, 'rb') as csvfile:
reader = db_action.get_table_data('status',file_id)
#reader = [dict(status_id='759985267390294_1158001970921953')]
for status in reader:
has_next_page = True
comments = getFacebookCommentFeedData(status['status_id'],
access_token, 100)
while has_next_page and comments is not None:
for comment in comments['data']:
db_action.save_to_db(processFacebookComment(comment,status['status_id']),'comments',file_id)
#w.writerow(processFacebookComment(comment,
# status['status_id']))
if 'comments' in comment:
has_next_subpage = True
subcomments = getFacebookCommentFeedData(
comment['id'], access_token, 100)
while has_next_subpage:
for subcomment in subcomments['data']:
# print (processFacebookComment(
# subcomment, status['status_id'],
# comment['id']))
db_action.save_to_db(processFacebookComment(
subcomment,
status['status_id'],
comment['id']),'comments',file_id)
num_processed += 1
if num_processed % 1000 == 0:
print "%s Comments Processed: %s" % \
(num_processed,
datetime.datetime.now())
if 'paging' in subcomments:
if 'next' in subcomments['paging']:
subcomments = json.loads(
request_until_succeed(
subcomments['paging']\
['next']))
else:
has_next_subpage = False
else:
has_next_subpage = False
# output progress occasionally to make sure code is not
# stalling
num_processed += 1
if num_processed % 1000 == 0:
print "%s Comments Processed: %s" % \
(num_processed, datetime.datetime.now())
if 'paging' in comments:
if 'next' in comments['paging']:
comments = json.loads(request_until_succeed(
comments['paging']['next']))
else:
has_next_page = False
else:
has_next_page = False
#db_action.close_db_connection()
print "\nDone!\n%s Comments Processed in %s" % \
(num_processed, datetime.datetime.now() - scrape_starttime)
if __name__ == '__main__':
print ('Invalid action! Please run "crawler_job.py" instead.')
# scrapeFacebookPageFeedComments(file_id, access_token)
# The CSV can be opened in all major statistical programs. Have fun! :)