def __init__(self, input_file='scrape_data.csv', config='options.ini'):
self.input_file = input_file
self.data_dict = DefaultOrderedDict(list)
ts = time()
self.stamp = datetime.fromtimestamp(ts).strftime('%Y-%m-%d %H:%M:%S')
self.read_csv()
options = os.path.join(os.path.dirname(__file__), config)
self.config = ConfigParser()
self.config.read(options)
def __init__(self, input_file='scrape_data.csv', config='options.ini'):
self.input_file = input_file
self.data_dict = DefaultOrderedDict(list)
ts = time()
self.stamp = datetime.fromtimestamp(ts).strftime('%Y-%m-%d %H:%M:%S')
self.read_csv()
options = os.path.join(os.path.dirname(__file__),config)
self.config = ConfigParser()
self.config.read(options)
class Social_Media_Tracker(object):
"""
Track a single entity across 3 diferent types of social media:
Facebook, Twitter and Youtube.
As of now you can retrieve:
-No. of likes from facebook
-No. of followers from Twitter
-No. of Subscribers from Youtube
This class takes a CSV input file with the following header:
twitter_id, facebook_id, youtube_id, handle, scraping frequency (times/hour),
likes, followers, subscribers, video_views.
It also has a configuration file (.ini) for FB and Twitter App keys.
The Youtube pinger method has dependencies that reside in current directory.
"""
def __init__(self, input_file='scrape_data.csv', config='options.ini'):
self.input_file = input_file
self.data_dict = DefaultOrderedDict(list)
ts = time()
self.stamp = datetime.fromtimestamp(ts).strftime('%Y-%m-%d %H:%M:%S')
self.read_csv()
options = os.path.join(os.path.dirname(__file__),config)
self.config = ConfigParser()
self.config.read(options)
def read_csv(self):
"""Loads input csv file into memory as dict"""
with open(self.input_file, 'r') as f:
self.data = csv.DictReader(f)
for row in self.data:
for key,value in row.iteritems():
self.data_dict[key].append(value)
def write_csv(self):
"""Writes back to csv file"""
with open('test.csv', 'wb') as f:
writer = csv.writer(f)
values = self.data_dict.values()
writer.writerow(self.data_dict.keys())
for vals in range(len(values[0])):
writer.writerow([row[vals] for row in values])
def log_error(self,error):
"""Logs error into txt file for debugging"""
with open('error_log.txt', 'a') as f:
f.write(str(error)+'\n')
def ping_facebook(self):
"""pings facebook for number of likes in current time"""
#Facebook API authentication procedure
token = self.config.get('facebook','token')
graph = facebook.GraphAPI(token)
#Look up number of likes from facebook_id written in csv
for i,ID in enumerate(self.data_dict['facebook_id']):
node = graph.get_object(ID)
try:
print 'facebook likes:', node['likes'], i
if i>0:
self.data_dict['likes '+self.stamp].append(node['likes'])
else:
self.data_dict['likes '+self.stamp] = [node['likes']]
except Exception, error:
self.log_error(error)
print self.data_dict
class Social_Media_Tracker(object):
"""
Track a single entity across 3 diferent types of social media:
Facebook, Twitter and Youtube.
As of now you can retrieve:
-No. of likes from facebook
-No. of followers from Twitter
-No. of Subscribers from Youtube
This class takes a CSV input file with the following header:
twitter_id, facebook_id, youtube_id, handle, scraping frequency (times/hour),
likes, followers, subscribers, video_views.
It also has a configuration file (.ini) for FB and Twitter App keys.
The Youtube pinger method has dependencies that reside in current directory.
"""
def __init__(self, input_file='scrape_data.csv', config='options.ini'):
self.input_file = input_file
self.data_dict = DefaultOrderedDict(list)
ts = time()
self.stamp = datetime.fromtimestamp(ts).strftime('%Y-%m-%d %H:%M:%S')
self.read_csv()
options = os.path.join(os.path.dirname(__file__), config)
self.config = ConfigParser()
self.config.read(options)
def read_csv(self):
"""Loads input csv file into memory as dict"""
with open(self.input_file, 'r') as f:
self.data = csv.DictReader(f)
for row in self.data:
for key, value in row.iteritems():
self.data_dict[key].append(value)
def write_csv(self):
"""Writes back to csv file"""
with open('test.csv', 'wb') as f:
writer = csv.writer(f)
values = self.data_dict.values()
writer.writerow(self.data_dict.keys())
for vals in range(len(values[0])):
writer.writerow([row[vals] for row in values])
def log_error(self, error):
"""Logs error into txt file for debugging"""
with open('error_log.txt', 'a') as f:
f.write(str(error) + '\n')
def ping_facebook(self):
"""pings facebook for number of likes in current time"""
#Facebook API authentication procedure
token = self.config.get('facebook', 'token')
graph = facebook.GraphAPI(token)
#Look up number of likes from facebook_id written in csv
for i, ID in enumerate(self.data_dict['facebook_id']):
node = graph.get_object(ID)
try:
print 'facebook likes:', node['likes'], i
if i > 0:
self.data_dict['likes ' + self.stamp].append(node['likes'])
else:
self.data_dict['likes ' + self.stamp] = [node['likes']]
except Exception, error:
self.log_error(error)
print self.data_dict
