def probYearGivenAgeFromFile(age, n_sample=0):
datadir = "../data"
filename = "ageEmbededTweets-Jun20.jsonarr"
names = []
for line in open(os.path.join(datadir, filename)):
tweet = json.loads(line)
if age == int(tweet["user"]["age"]):
name = tweet["user"]["name"]
names.append(friends.firstNameOf(name))
if n_sample != 0 and n_sample < len(names):
names = random.sample(names, n_sample)
return probYearGivenNames(names)
def insertUsersIntoMySQL():
"""
Read the users data retrieved from Twitter. The retrieval is done by
crawler.py. The mysql db will be:
(rowid, screen_name, user_id, age, name, eng)
The table might be created by using the below sql.
CREATE TABLE `users` (
`rowid` bigint(20) NOT NULL AUTO_INCREMENT,
`screen_name` varchar(20) NOT NULL,
`user_id` bigint(20) NOT NULL,
`age` int(11) DEFAULT NULL,
`name` varchar(50) DEFAULT NULL,
`eng` tinyint(1) DEFAULT NULL,
PRIMARY KEY (`rowid`),
UNIQUE KEY `user_id` (`user_id`),
KEY `eng` (`eng`)
) ENGINE=MyISAM DEFAULT CHARSET=utf8;
"""
filedir = "/home/pyongjoo/workspace/twitter-research/data"
filename = "users_Jun20_step0-1_2.1.jsonarr"
f = open(os.path.join(filedir, filename))
con = db.con()
with con:
cur = con.cursor()
for line in f:
j = json.loads(line) # object read using json
screen_name = j["screen_name"]
user_id = j["id"]
name = friends.firstNameOf(j["name"])
eng = 1 if j["lang"] == "en" else 0
cur.execute(
"""INSERT INTO users VALUES
(0, %s, %s, NULL, %s, %s)""",
(screen_name, user_id, name, eng),
)
f.close()
