def main():
controlTweets = None
cohortTweets = None
with dbUtils.setup_mysql_cxn() as cxn:
curs = cxn.cursor()
curs.execute(SELECT_CONTROL_TWEETS)
controlTweets = dbUtils.get_named_rows(curs)
curs.execute(SELECT_COHORT_TWEETS)
cohortTweets = dbUtils.get_named_rows(curs)
make_cdfs("control", controlTweets)
make_cdfs("cohort", cohortTweets)
def write_csv (fname, results):
fields = list(results[0].keys())
fields.sort()
with open(fname, "w+") as f_out:
writer = csv.writer(f_out, delimiter="\t")
writer.writerow(fields)
for row in results:
row_elems = [row[k] if k in row else 0 for k in fields]
writer.writerow(row_elems)
if __name__ == "__main__":
processor = TweetProcessor()
results = []
with dbUtils.setup_mysql_cxn() as cxn:
curs = cxn.cursor();
curs.execute(dbUtils.GET_USERS_W_COHORT)
users = dbUtils.get_named_rows(curs)
for user in users:
curs.execute(dbUtils.GET_TWEETS_TEMPL % user['twitter_user_id'])
tweets = dbUtils.get_named_rows(curs)
result = defaultdict(lambda: 0)
for tweet in tweets:
result = processor.process_tweet(tweet['tweet_text'], result)
ntweets = len(tweets)
if ntweets > 0:
for k, v in result.items():
result[k] = float(v) / float(ntweets)
result = add_time_model(tweets, result)
