def get_user_data(user_id=None, screen_name=None):
"""
For a given user, get user profile data
Keyword args:
`screen_name` or `user_id` (not both!): either user screen_name (@potus) or a user id (110123231)
---
cf https://dev.twitter.com/rest/reference/get/users/show
"""
# Must have either screen_name or user_id (logical xor)
assert (screen_name != None) != (user_id != None),\
"Must have screen_name or user_id, but not both"
twitter_api = conn.get_twitter_api()
get_user_data = partial(
make_twitter_request,
twitter_api.show_user)
if user_id:
user_data = get_user_data(user_id=user_id)
else:
user_data = get_user_data(screen_name=screen_name)
return user_data
def get_friends_followers_ids(
screen_name=None, user_id=None,
friends_limit=maxsize, followers_limit=maxsize):
"""
For a given user, get list of followers and friends ids
Keyword args:
`twitter_api`: a twython.Twython authentication object
`screen_name` or `user_id` (not both!): either user screen_name (@potus) or a user id (110123231)
`friends_limit`: Max number of friends to return (set to 0 if only interested in getting followers)
`followers_limit`: Max number of friends to return (set to 0 if only interested in getting friends)
---
cf https://dev.twitter.com/rest/reference/get/friends/ids
cf https://dev.twitter.com/rest/reference/get/followers/ids
"""
# Must have either screen_name or user_id (logical xor)
assert (screen_name != None) != (user_id != None),\
"Must have screen_name or user_id, but not both"
twitter_api = conn.get_twitter_api()
get_followers_ids = partial(
make_twitter_request,
twitter_api.get_followers_ids,
count=5000)
get_friends_ids = partial(
make_twitter_request,
twitter_api.get_friends_ids,
count=5000)
friends_ids, followers_ids = [], []
for twitter_api_func, limit, ids, label in [
[get_friends_ids, friends_limit, friends_ids, "friends"],
[get_followers_ids, followers_limit, followers_ids, "followers"]
]:
if limit == 0:
continue
cursor = -1
while cursor != 0:
# Use make_twitter_request via the partially bound callable...
if screen_name:
response = twitter_api_func(screen_name=screen_name, cursor=cursor)
else: # user_id
response = twitter_api_func(user_id=user_id, cursor=cursor)
if response is not None:
ids.extend(response['ids'])
cursor = response['next_cursor']
print ('Fetched {0} total {1} ids for {2}'.format(len(ids),
label, (user_id or screen_name)))
# XXX: You may want to store data during each iteration to provide
# an additional layer of protection from exceptional circumstances
if len(ids) >= limit or response is None:
break
# Do something useful with the IDs, like store them to disk...
return friends_ids[:friends_limit], followers_ids[:followers_limit]
def search(q=None,
since_id=None,
max_id=None,
tweet_limit=sys.maxsize,
save_dir=None):
"""
For a given query q, get recent tweets (up to 7 days old)
Keyword args:
`q`: keyword query
`since_id`: id of the most ancient tweets to fetch up to.
`max_id`: id of the most recent tweet to start fetching from (get timeline goes backward in time when fetching tweets)
`tweet_limit`: Max number of tweets to return
---
https://developer.twitter.com/en/docs/tweets/search/api-reference/get-search-tweets.html
for details on API parameters
"""
tweets = []
twitter_api = conn.get_twitter_api()
search_tweets = partial(make_twitter_request, twitter_api.search)
cursor = max_id
while cursor != 0:
response = search_tweets(q=q,
since_id=since_id,
max_id=cursor,
count=100)
if response is None or response == []:
break
else:
try:
cursor = min([t['id'] for t in response['statuses']])
except:
print("'response' does not contain a 'statuses' key:")
print(list(response.keys()))
new_tweets = [t for t in response['statuses'] if t not in tweets]
if new_tweets:
tweets.extend(new_tweets)
print("max_id = {cursor}".format(cursor=cursor))
if save_dir is not None:
for tweet in new_tweets:
save_json(tweet, save_dir)
else:
break
return tweets
print("max_id = {cursor}".format(cursor=cursor))
if save_dir is not None:
for tweet in new_tweets:
save_json(tweet, save_dir)
else:
break
return tweets
if __name__ == "__main__":
import sys
assert len(sys.argv) == 3, """
Call not properly formatted;
Correct format:
'''
python3 search.py {search_term} {save_dir}
'''
"""
search_term = sys.argv[1]
save_dir = sys.argv[2]
import conn
twitter_api = conn.get_twitter_api()
try:
tweets = search(q=search_term, save_dir=save_dir)
except TwythonError as e:
print(e)
def get_user_timeline(
screen_name=None, user_id=None,
since_id=None,
max_id=None,
tweet_limit=maxsize):
"""
For a given user, get user_timeline
Keyword args:
`twitter_api`: a twython.Twython authentication object
`screen_name` or `user_id` (not both!): either user screen_name (@potus) or a user id (110123231)
`max_id`: id of the most recent tweet to start fetching from (get timeline goes backward in time when fetching tweets)
`tweet_limit`: Max number of tweets to return
---
https://dev.twitter.com/rest/reference/get/statuses/user_timeline
"""
# Must have either screen_name or user_id (logical xor)
assert (screen_name != None) != (user_id != None),\
"Must have screen_name or user_id, but not both"
# https://dev.twitter.com/rest/reference/get/statuses/user_timeline
# for details on API parameters
twitter_api = conn.get_twitter_api()
get_user_timeline = partial(
make_twitter_request,
twitter_api.get_user_timeline)
user_timeline = []
cursor = max_id
while cursor != 0:
# Use make_twitter_request via the partially bound callable...
if screen_name:
response = get_user_timeline(screen_name=screen_name,
since_id=since_id,
include_rts=True,
max_id=cursor,
count=200)
user_identifier = screen_name # identifier for the user
else: # user_id
response = get_user_timeline(user_id=user_id,
since_id=since_id,
include_rts=True,
max_id=cursor,
count=200)
user_identifier = user_id # identifier for the user
if response is None or response==[]:
break
else:
ids = [t['id'] for t in response]
cursor = min(ids)
new_tweets = [t for t in response if t not in user_timeline]
if new_tweets:
user_timeline.extend(new_tweets)
print("max_id = {cursor}".format(cursor=cursor))
else:
break
return user_timeline