def arquive_search(self,
query,
start,
end,
dev_env,
max_size=2500,
max_call=100):
self.settings['search_tweets_api']['endpoint'] =\
f"https://api.twitter.com/1.1/tweets/search/fullarchive/{dev_env}.json"
credentials = load_credentials("archive_keys.yaml",
yaml_key="search_tweets_api",
env_overwrite=False)
with open('archive_keys.yaml', 'w') as config_file:
yaml.dump(self.settings, config_file, default_flow_style=False)
q_rule = gen_rule_payload(query,
results_per_call=max_call,
from_date=start,
to_date=end)
rs = ResultStream(rule_payload=q_rule,
max_results=max_size,
**credentials)
with open('tweet_data_archive.csv', 'a', encoding='utf-8') as file:
n = 0
for tweet in rs.stream():
n += 1
if n % (max_size / 10) == 0:
print('{0}: {1}'.format(str(n), tweet['created_at']))
json.dump(tweet, file)
file.write('\n')
def main():
args_dict = vars(parse_cmd_args().parse_args())
if args_dict.get("debug") is True:
logger.setLevel(logging.DEBUG)
logger.debug("command line args dict:")
logger.debug(json.dumps(args_dict, indent=4))
if args_dict.get("config_filename") is not None:
configfile_dict = read_config(args_dict["config_filename"])
else:
configfile_dict = {}
extra_headers_str = args_dict.get("extra_headers")
if extra_headers_str is not None:
args_dict['extra_headers_dict'] = json.loads(extra_headers_str)
del args_dict['extra_headers']
logger.debug("config file ({}) arguments sans sensitive args:".format(
args_dict["config_filename"]))
logger.debug(json.dumps(_filter_sensitive_args(configfile_dict), indent=4))
creds_dict = load_credentials(filename=args_dict["credential_file"],
account_type=args_dict["account_type"],
yaml_key=args_dict["credential_yaml_key"],
env_overwrite=args_dict["env_overwrite"])
dict_filter = lambda x: {k: v for k, v in x.items() if v is not None}
config_dict = merge_dicts(dict_filter(configfile_dict),
dict_filter(creds_dict), dict_filter(args_dict))
logger.debug("combined dict (cli, config, creds) sans password:"******"ERROR: not enough arguments for the program to work")
sys.exit(1)
stream_params = gen_params_from_config(config_dict)
logger.debug(
"full arguments passed to the ResultStream object sans password")
logger.debug(json.dumps(_filter_sensitive_args(stream_params), indent=4))
rs = ResultStream(tweetify=False, **stream_params)
logger.debug(str(rs))
if config_dict.get("filename_prefix") is not None:
stream = write_result_stream(
rs,
filename_prefix=config_dict.get("filename_prefix"),
results_per_file=config_dict.get("results_per_file"))
else:
stream = rs.stream()
for tweet in stream:
if config_dict["print_stream"] is True:
print(json.dumps(tweet))
def collect_tweets_in_files():
""" Using a ResultStream for getting tweets
We can configure the amount of pages/tweets we want to obtain """
if not check_files(): # file should not already be existing
max_results = 10000
max_pages = 300
max_tweets = 15000
rs = ResultStream(request_parameters=query,
max_results=max_results,
max_pages=max_pages,
**credentials)
# Set how many tweets we want to catch
rs.max_tweets = max_tweets
tweets_2 = list(rs.stream())
dataframe = pandas.DataFrame(tweets_2)
csv_file = dataframe.to_csv(saving_path)
else:
print(
FileExistsError,
'File already exists! Please check if you really want to overwrite the file.'
)
def search(queryString, outputpath, api_key_yaml,startTime="2016-01-01",endTime="2021-03-15", lang="en"):
search_args = load_credentials(api_key_yaml,
yaml_key="search_tweets_v2",
env_overwrite=False)
print("Should be 1024, but it:")
print(len(queryString + " -is:nullcast -is:retweet -is:verified -is:quote " + "lang:"+lang))
#,created_at,geo,in_reply_to_user_id,lang,author_id,conversation_id,public_metrics,entities,context_annotations
query = gen_request_parameters(query=queryString.strip() + " -is:nullcast -is:retweet -is:verified -is:quote " + "lang:"+lang, media_fields="media_key,type",user_fields="id,description,location,name,entities,url,username,public_metrics,verified,withheld,protected",tweet_fields="id,text,created_at,geo,in_reply_to_user_id,lang,author_id,conversation_id,public_metrics,entities,context_annotations,attachments",start_time=startTime,end_time=endTime, stringify=False, expansions="author_id,attachments.media_keys",results_per_call=500)
rs = ResultStream(request_parameters=query, max_tweets=sys.maxsize, max_requests=sys.maxsize, **search_args)
i = 0
with open(outputpath, 'w') as outputcsv:
writer = csv.writer(outputcsv)
writer.writerow(headers)
for tweet in rs.stream():
# print(tweet)
if "id" in tweet:
writer.writerow(createRow(headers, tweet))
if "users" in tweet:
print("parsing users")
dump_users_info(tweet,outputpath.replace(".csv",str(i) +"-users.csv"))
i+=1
def gather_data(self, screen_name: str, user_id: int, rt_date: str, file_path: str):
query_str = create_query_str(screen_name)
# print(f'reconstructing timeline for @{screen_name}')
time_range = get_start_and_end_date(rt_date)
query_obj = create_query_obj(query_str, *time_range)
rs = ResultStream(
request_parameters=query_obj,
# parameter changed from 2 -> 1 to avoid being ratelimited within the project timeline
max_requests=1,
**self.academic_search_args
)
inbound_timeline = []
replies = []
retweets = []
quotes = []
for tweet in rs.stream():
if "author_id" not in tweet:
if "tweets" in tweet:
# Tweets are found
for t in tweet["tweets"]:
if int(t["author_id"]) == user_id:
if "referenced_tweets" in t:
ref_tweets = t["referenced_tweets"]
for ref in ref_tweets:
type = ref["type"]
if type == "replied_to":
replies.append(ref["id"])
elif type == "quoted":
quotes.append(ref["id"])
else:
# normal tweet, which holds no info on the information strength
pass
else:
if "referenced_tweets" not in t:
# the only way this situation can occur is when the tweet is retweeted by the autor
# and someone is replying to that retweet
retweets.append(t["author_id"])
else:
# this indicates a reply with a quote, or a reply of a reply
pass
# print(f"done collecting the retweeted user objects, there are {len(retweets)} in total")
# print(f"converting the {len(replies)} replied tweet objects to user ids")
replies = self.gather_users(replies)
# print(f"done collecting the replies user objects, there are {len(replies)} in total")
# print(f"converting the {len(quotes)} quoted tweet objects to user ids")
quotes = self.gather_users(quotes)
# print(f"done collecting the quotes user objects, there are {len(quotes)} in total")
# print(f"retweets: {len(retweets)}\treplies: {len(replies)}\tquotes: {len(quotes)}")
dump_dict = {"replies": replies, "quotes": quotes, "retweets": retweets}
json.dump(dump_dict, open(file_path, "w"))
def _download_tweets(trend, enterprise_search_args):
powertrack_rule = '(has:geo OR has:profile_geo) lang:en -is:retweet %s' % trend
rule = gen_rule_payload(powertrack_rule, results_per_call=500)
rs = ResultStream(rule_payload=rule,
max_requests=2,
**enterprise_search_args)
for tweet in rs.stream():
print(tweet)
_store_tweet(tweet)
def usersTweetsByIds():
search_args1 = load_credentials(".twitter_keys.yaml",
yaml_key="search_tweets_v2_id",
env_overwrite=False)
search_args2 = load_credentials(".twitter_keys.yaml",
yaml_key="search_tweets_v2_user",
env_overwrite=False)
f = open(
'C:\\Users\\Josh\\Documents\\GitHub\\search-tweets-python\\enUsers_Tweets.json',
'r',
encoding='utf-8')
obj = json.load(f)
for u in obj['includes']:
idList = u.get('tweetids')
ids = ''
idList = list(set(idList))
if len(idList) == 0:
u['tweets'] = []
continue
if len(idList) > 99:
ids = ','.join(idList[0:99])
else:
ids = ','.join(idList)
endTweet = 'https://api.twitter.com/2/tweets'
query = {"ids": ids, "tweet.fields": "author_id,public_metrics,text"}
rs = ResultStream(request_parameters=query,
endpoint=endTweet,
bearer_token=bt)
tweets = []
result = list(rs.stream())
for r in result:
tweets = r.get('data')
u['tweets'] = tweets
fo = open('Random_WithTweets.json', 'w', encoding='utf-8')
json.dump(obj, fo)
def _download_tweets(trend):
powertrack_rule = '%s (has:geo OR has:profile_geo) lang:en -is:retweet' % trend
rule = gen_rule_payload(powertrack_rule,
results_per_call=500,
to_date=None,
from_date='201207220000')
logging.info("PowerTrack rule: %s" % rule)
rs = ResultStream(rule_payload=rule,
max_results=500,
max_requests=1,
**enterprise_search_args)
for tweet in rs.stream():
_push_tweet(tweet, trend)
def main():
parser = parse_cmd_args()
args_dict = vars(parse_cmd_args().parse_args())
if args_dict.get("debug") is True:
logger.setLevel(logging.DEBUG)
logger.debug(json.dumps(args_dict, indent=4))
if args_dict.get("config_filename") is not None:
configfile_dict = read_config(args_dict["config_filename"])
else:
configfile_dict = {}
creds_dict = load_credentials(filename=args_dict["credential_file"],
account_type=args_dict["account_type"],
yaml_key=args_dict["credential_yaml_key"],
env_overwrite=args_dict["env_overwrite"])
dict_filter = lambda x: {k: v for k, v in x.items() if v is not None}
config_dict = merge_dicts(dict_filter(configfile_dict),
dict_filter(args_dict), dict_filter(creds_dict))
logger.debug(json.dumps(config_dict, indent=4))
if len(dict_filter(config_dict).keys()
& REQUIRED_KEYS) < len(REQUIRED_KEYS):
print(REQUIRED_KEYS - dict_filter(config_dict).keys())
logger.error("ERROR: not enough arguments for the program to work")
sys.exit(1)
stream_params = gen_params_from_config(config_dict)
logger.debug(json.dumps(config_dict, indent=4))
rs = ResultStream(tweetify=False, **stream_params)
logger.debug(str(rs))
if config_dict.get("filename_prefix") is not None:
stream = write_result_stream(
rs,
filename_prefix=config_dict["filename_prefix"],
results_per_file=config_dict["results_per_file"])
else:
stream = rs.stream()
for tweet in stream:
if config_dict["print_stream"] is True:
print(json.dumps(tweet))
def tw_get_premium_search(self, keyword: str):
with open(f'datasets/tw_{keyword.lower()}_searches_premium.json',
'w') as f:
try:
f.write('{"statuses": [')
rule = gen_rule_payload(
pt_rule="near:\"New York, NY\" within:50mi".format(),
results_per_call=100,
from_date="2018-07-01",
to_date="2018-10-01")
rule = gen_rule_payload(
pt_rule="place:\"New York, NY\"".format(),
results_per_call=100,
from_date=(datetime.date.today() -
datetime.timedelta(31)).isoformat(),
to_date=datetime.date.today().isoformat())
next_token = None
while True:
results = ResultStream(rule_payload=rule,
**self.twitter_premium_api)
results.next_token = next_token
tweets = []
try:
tweets = list(results.stream())
except Exception as ex:
print(str(ex))
for tweet in tweets:
f.write("%s," % json.dumps(tweet))
if results.next_token is None:
break
else:
next_token = results.next_token
next_token is not None and f.seek(f.tell() - 1, os.SEEK_SET)
f.write("]}")
except Exception as ex:
print("Error:\n" + str(ex))
def get_data(search_query, api_key, secret_key, to_date, from_date, filename):
""" get twitter data through twitter API from full archive search sand box and return all twitters in JSONL file
based on
search term,
the geographic location of interest
the time period of interest.
and personal twitter account information.
Reference: https://github.com/geduldig/TwitterAPI/tree/master/TwitterAPI
Reference: https://developer.twitter.com/en/docs/tweets/search/overview
"""
print_after_x = 1000
config = dict(
search_tweets_api=dict(
account_type='premium',
endpoint=f"https://api.twitter.com/1.1/tweets/search/{'fullarchive'}/{'mangroveConservation'}.json",
consumer_key=api_key,
consumer_secret=secret_key
)
)
with open('twitter_keys.yaml', 'w') as config_file:
yaml.dump(config, config_file, default_flow_style=False)
from searchtweets import load_credentials, gen_rule_payload, ResultStream
premium_search_args = load_credentials("twitter_keys.yaml",
yaml_key="search_tweets_api",
env_overwrite=False)
rule = gen_rule_payload(search_query,
results_per_call=100,
from_date=from_date,
to_date=to_date
)
temp = ResultStream(rule_payload=rule,
max_results=100000,
**premium_search_args)
with open(filename, 'a', encoding='utf-8') as temp_file:
num = 0
for tweet in temp.stream():
num += 1
if num % print_after_x == 0:
print('{0}: {1}'.format(str(num), tweet['created_at']))
json.dump(tweet, temp_file)
temp_file.write('\n')
print('done')
def save_old_tweets():
from searchtweets import load_credentials, gen_rule_payload, ResultStream
import json
premium_search_args = load_credentials("twitter_keys_fullarchive.yaml",
yaml_key="search_tweets_api",
env_overwrite=False)
query = "from:NTOO_Org"
rule = gen_rule_payload(query, results_per_call=100)
rs = ResultStream(rule_payload=rule,
max_results=1000,
**premium_search_args)
with open('fullTweetsData.json', 'a', encoding='utf-8') as f:
for tweet in rs.stream():
json.dump(tweet, f)
f.write('\n')
def tweet_search(search_key, search_args):
"""
search for "spectrumtv" and create a dict of tweet timestamp (dictionary key, in epoch seconds),
tweet authors screen name (dict value, tuple element 1),
tweet text (dict value, tuple element 2)
"""
print("searching for tweets containing \"{}\"".format(search_key))
key_rule = gen_rule_payload(search_key, results_per_call=100)
key_rs = ResultStream(rule_payload=key_rule,
max_results=500,
max_pages=1,
**search_args)
key_results = list(key_rs.stream())
key_tweets = {}
for tweet in key_results:
key_tweets[tweet.created_at_seconds] = (
tweet.screen_name, tweet.all_text.replace('\n', ' '), ' '
) # this space is a placeholder for the sentiment value
print("{} tweets found containing \"{}\"\n".format(len(key_results),
search_key))
return key_tweets
def pull_data_for_handle(self,
handle,
date,
days_before,
results_per_call=100,
max_results=2500):
# check handle can be found!
user_id = self.get_handle_id(handle)
if user_id is 0:
return 0
from_date = self.subtract_from_datestring(date, days_before)
rule = self.make_rule(handle, date, from_date, results_per_call)
rs = ResultStream(rule_payload=rule,
max_results=max_results,
**self.endpoint_args)
results_list = list(rs.stream())
# results_list=temp_dict[list(temp_dict.keys())[0]]
print('Found', len(results_list), 'tweets for', handle)
if len(results_list) == max_results:
print('Max results limit hit (' + str(2500) +
'). Consider changing the parameter')
return self.strip_maxresults_from_query(rule), results_list
def getRecentTweets():
endRecent = 'https://api.twitter.com/2/tweets/search/recent'
search_args_rec = load_credentials(".twitter_keys.yaml",
yaml_key="search_tweets_v2_recent",
env_overwrite=False)
query = {
"max_results": 100,
"tweet.fields": "public_metrics,author_id,lang",
"query":
"happy -RT OR upset -RT OR lol -RT OR ugh -RT OR dog -RT OR cat -RT OR food -RT OR sucks -RT",
"expansions": "author_id",
"user.fields": "public_metrics"
}
rs = ResultStream(
request_parameters=query,
endpoint=endRecent,
bearer_token=bt,
max_tweets=100,
max_requests=1,
)
result = list(rs.stream())
obj = {}
obj['data'] = []
obj['includes'] = []
for r in result:
obj['data'] = obj['data'] + r.get('data')
obj['includes'] = obj['includes'] + r.get('includes').get('users')
out = open('testJson.json', 'w')
json.dump(obj, out)
def get_twitter_results(news_id,
query,
from_date,
premium_search_args,
filename,
to_date="202005260000"):
query1 = "url:" + query + " lang:en"
rule = gen_rule_payload(query1,
from_date=from_date,
to_date=to_date,
results_per_call=100)
rs = ResultStream(rule_payload=rule,
max_results=100,
**premium_search_args)
l = 0
with open(filename, 'a', encoding='utf-8') as f:
n = 0
for tweet in rs.stream():
news_tweet_json = {
"news_id": news_id,
"query": query,
"tweet": tweet
}
n += 1
if n % 10 == 0:
print('{0}: {1}'.format(str(n), tweet['created_at']))
json.dump(news_tweet_json, f)
f.write('\n')
l = datetime.strptime(tweet['created_at'],
"%a %b %d %H:%M:%S +%f %Y").date()
print(rs, type(l), l)
print('done')
return l
def main():
args_dict = vars(parse_cmd_args().parse_args())
if args_dict.get("debug") is True:
logger.setLevel(logging.DEBUG)
logger.debug("command line args dict:")
logger.debug(json.dumps(args_dict, indent=4))
if args_dict.get("config_filename") is not None:
configfile_dict = read_config(args_dict["config_filename"])
else:
configfile_dict = {}
extra_headers_str = args_dict.get("extra_headers")
if extra_headers_str is not None:
args_dict['extra_headers_dict'] = json.loads(extra_headers_str)
del args_dict['extra_headers']
logger.debug("config file ({}) arguments sans sensitive args:".format(
args_dict["config_filename"]))
logger.debug(json.dumps(_filter_sensitive_args(configfile_dict), indent=4))
creds_dict = load_credentials(filename=args_dict["credential_file"],
yaml_key=args_dict["credential_yaml_key"],
env_overwrite=args_dict["env_overwrite"])
dict_filter = lambda x: {k: v for k, v in x.items() if v is not None}
config_dict = merge_dicts(dict_filter(configfile_dict),
dict_filter(creds_dict), dict_filter(args_dict))
logger.debug("combined dict (cli, config, creds):")
logger.debug(json.dumps(_filter_sensitive_args(config_dict), indent=4))
if len(dict_filter(config_dict).keys()
& REQUIRED_KEYS) < len(REQUIRED_KEYS):
print(REQUIRED_KEYS - dict_filter(config_dict).keys())
logger.error("ERROR: not enough arguments for the script to work")
sys.exit(1)
stream_params = gen_params_from_config(config_dict)
logger.debug(
"full arguments passed to the ResultStream object sans credentials")
logger.debug(json.dumps(_filter_sensitive_args(stream_params), indent=4))
while True:
start = time.time()
rs = ResultStream(tweetify=False, **stream_params)
logger.debug(str(rs))
if config_dict.get("filename_prefix") is not None:
stream = write_result_stream(
rs,
filename_prefix=config_dict.get("filename_prefix"),
results_per_file=config_dict.get("results_per_file"))
else:
stream = rs.stream()
first_tweet = True
tweets_num = 0
#Iterate through Tweet array and handle output.
for tweet in stream:
tweets_num = tweets_num + 1
#Get Tweet ID from first Tweet
if first_tweet:
newest_id = tweet['id']
first_tweet = False
if config_dict["print_stream"] is True:
print(json.dumps(tweet))
#This polling script switches to a since_id requests and removes the start_time parameter if it is used for backfill.
#Prepare next query, by setting the since_id request parameter.
print(f"{tweets_num} new Tweets. Newest_id: {newest_id}")
request_json = json.loads(stream_params['request_parameters'])
if 'start_time' in request_json.keys():
del request_json['start_time']
request_json.update(since_id=newest_id)
stream_params['request_parameters'] = json.dumps(request_json)
duration = time.time() - start
sleep_interval = (float(config_dict["interval"]) * 60) - duration
if sleep_interval < 0:
sleep_interval = (float(config_dict["interval"]) * 60)
time.sleep(sleep_interval)
def extract_tweets():
today = date.today()
d1 = today.strftime("%d-%m-%Y")
with open('config.json','r') as f:
keys = json.load(f)
config = dict(
search_tweets_api = dict(
account_type = 'premium',
endpoint = 'https://api.twitter.com/1.1/tweets/search/30day/development1.json',
consumer_key = keys['consumer_key'],
consumer_secret = keys['consumer_secret'])
)
with open('twitter_keys_fullhistory.yaml', 'w') as config_file:
yaml.dump(config, config_file, default_flow_style=False)
premium_search_args = load_credentials("twitter_keys_fullhistory.yaml",
yaml_key="search_tweets_api",
env_overwrite=False)
SEARCH_QUERY = 'to:Lloydsbank'
RESULTS_PER_CALL = 100
FROM_DATE = "2020-06-01"
TO_DATE = "2020-06-10"
MAX_RESULTS = 100000
FILENAME = 'twitter_input_data_{}_{}.jsonl'.format(FROM_DATE, TO_DATE) # Where the Tweets should be saved
PRINT_AFTER_X = 100
rule = gen_rule_payload(SEARCH_QUERY,
results_per_call=RESULTS_PER_CALL,
from_date=FROM_DATE,
to_date=TO_DATE
)
rs = ResultStream(rule_payload=rule,
max_results=MAX_RESULTS,
**premium_search_args)
with open(FILENAME, 'a', encoding='utf-8') as f:
n = 0
for tweet in rs.stream():
n += 1
if n % PRINT_AFTER_X == 0:
print('{0}: {1}'.format(str(n), tweet['created_at']))
json.dump(tweet, f)
f.write('\n')
new_tweets = []
dates_created = []
location = []
user= []
with open(FILENAME, 'rb') as f:
for item in json_lines.reader(f):
try:
new_tweets.append(item['extended_tweet']['full_text'])
except KeyError as e:
new_tweets.append(item['text'])
dates_created.append(item['created_at'])
location.append(item['user']['location'])
user.append(item['user']['id'])
dataframe = pd.DataFrame(list(zip(user, location, dates_created, new_tweets)),
columns =['User', 'Location', 'date_created', 'text'])
print(dataframe.head())
dataframe.to_csv("tweets.csv", sep =",")
from searchtweets import gen_rule_payload
from searchtweets import load_credentials
premium_search_args = load_credentials("twitter_keys_fullarchive.yaml",
yaml_key="search_tweets_api",
env_overwrite=False)
print(premium_search_args)
query = "AAPL"
rule = gen_rule_payload(query,
results_per_call=100,
from_date="2020-10-21",
to_date="2020-10-28")
from searchtweets import ResultStream
rs = ResultStream(rule_payload=rule, max_results=25000, **premium_search_args)
print(rs)
import json
with open('tweets1.json', 'a', encoding='utf-8') as f:
for tweet in rs.stream():
json.dump(tweet, f)
f.write('\n')
print('done')
max_tweets=config['max_tweets'],
**search_creds)
# number of reconnection tries
tries = 10
# while loop to protect against 104 error
while True:
tries -= 1
# attempt retrieving tweets
try:
# indicate which day is getting retrieved
print('[INFO] - Retrieving tweets from ' + str(start_ts))
# get json response to list
tweets = list(rs.stream())
# break free from while loop
break
except Exception as err:
if tries == 0:
raise err
else:
print(
'[INFO] - Got connection error, waiting 15 seconds and trying again. '
+ str(tries) + ' tries left.')
time.sleep(15)
# parse results to dataframe
print('[INFO] - Parsing tweets from ' + str(start_ts))
tweetdf = v2parser(tweets, config['results_per_call'])
def write_stream(self):
""" write ResultStream object to disk using the write_ndjson utility """
stream = ResultStream(**self.premium_search_args, rule_payload=self.rule, max_results=62000)
columns = []
for _ in write_ndjson('US_apr02_apr09_some.json', stream.stream()): # exhaust generator
pass
start_date = today + datetime.timedelta(-30)
print(start_date)
rule = gen_rule_payload("from:NYCASP",
from_date=str(start_date),
to_date=str(today),
results_per_call=500)
print(rule)
rs = ResultStream(rule_payload=rule, max_results=500, **premium_search_args)
print(rs)
tweets = rs.stream()
list_tweets = list(tweets)
[print(tweet.all_text, end="\n\n") for tweet in list_tweets[0:100]]
tweet_text = []
tweet_date = []
for tweet in list_tweets:
tweet_text.append(tweet["text"])
tweet_date.append(tweet["created_at"])
df = pd.DataFrame({"tweet": tweet_text, "date": tweet_date})
df.head()
client = twilio_connect()
def pull_tweets(query,
from_date,
to_date,
save_path,
credentials_path,
yaml_key,
file_name=None,
results_per_call=500,
max_results=3000,
verbose=False,
**kwargs):
"""
Pulls data (i.e., tweets and user info) from Twitter using its API.
The data received from the API is stored in its original form (JSON)
without performing any type of preprocessing.
Parameters
----------
query : str
Query passed to the Twitter API to fecth Tweets.
from_date : str or None
Date format as specified by `convert_utc_time` for the starting time
of your search.
to_date : str or None
Date format as specified by `convert_utc_time` for the end time of
your search.
save_path : str
Path where the raw data will be stored.
credentials_path : str
Path for the yaml file with the Twitter API credentials.
yaml_key : str
Key within the yaml file containing the Twitter API credentials to be
used.
file_name : str or None, default=None
Name of the json file saved containing the data dump. If None, the
named will be assigned as a function of `query`, `from_date` and
`to_date`.
results_per_call : int, default=500
Number of Tweets returned per call.
max_results : int, default=3000
Maximum number of Tweets to be pulled.
verbose : int or bool, default=False
Controls the verbosity when pulling data.
Returns
-------
None : NoneType
"""
logger = logging.getLogger(__name__)
logger.propagate = verbose
logger.info('Pulling raw Twitter data')
search_args = load_credentials(filename=credentials_path,
yaml_key=yaml_key)
rule = gen_rule_payload(query,
results_per_call=results_per_call,
from_date=from_date,
to_date=to_date)
rs = ResultStream(rule_payload=rule,
max_results=max_results,
**search_args)
if file_name is None:
file_name = f'SAMPLE_DATA_QUERY_{query}_'\
+ f'FROMDATE_{from_date}_TODATE_{to_date}.json'
with open(os.path.join(save_path, file_name), 'a', encoding='utf-8') as f:
for tweet in rs.stream():
json.dump(tweet, f)
f.write('\n')
logger.info('Data successfuly saved at' +
f'\"{os.path.join(save_path, file_name)}\"')
return None