"--user",
dest="user",
help=u"User (screen name in Twitter)",
metavar="USER",
default='')
(options, args) = parser.parse_args()
if not (options.input and options.iteractions and options.output
and options.type):
parser.print_usage()
exit(1)
(user_relevance, content_relevance, content_to_user,
user_photo) = compute_relevance(options.input, int(options.iteractions),
float(options.damping_factor),
options.user)
#(user_num_retweets,user_average_num_retweets,user_num_retweeters,user_sum_num_retweets_retweeter) = compute_user_statistics(input_file_name)
user_relevance_table = {}
with open(options.output, 'w') as output_file:
#output_file.write("user,influence,retweets,avg retweets,retweeters,sum retweeters\n")
for user in xrange(0, len(user_relevance)):
user_relevance_table[user_relevance[user]
[0]] = user_relevance[user][1]
if options.type == "USER":
def main():
parser = OptionParser()
# parser.add_option("-i", "--input", dest="input", help=u"Database input file (sqlite3)", metavar="INPUT")
parser.add_option("-i", "--input-dir", dest="input_dir",
help=u"Input dir containing CSV files."
u"Files must contain 1 line per tweet and fields username, tweet.id, retweet.id, "
u"tweet.create_date (yyyy-mm-ddThh:MM:ss), entities (comma-delimited, optional) and "
u"user.foto. File names must be in the format yyyymmdd. ",
metavar="INPUT")
parser.add_option("-n", "--iteractions", dest="iteractions", help=u"Number of iteractions (default = 20)",
metavar="ITERACTIONS", default=20)
parser.add_option("-d", "--damping-factor", dest="damping_factor", help=u"Damping factor (default: 0.85)",
metavar="DAMPING", default='0.85')
parser.add_option("-o", "--output", dest="output",
help=u"Output dir. Files will be named after type (CONTENT or USER) and date",
metavar="OUPUT")
parser.add_option("-t", "--type", dest="type", help=u"Type (USER or CONTENT)", metavar="TYPE")
parser.add_option("-u", "--user", dest="user", help=u"User (screen name in Twitter). Not used anymore.",
metavar="USER", default='')
parser.add_option("", "--date", dest="date", help=u"Reference date. Defaults to today, format yyyymmdd",
metavar="Date")
parser.add_option("-r", "--days", dest="days",
help=u"Number of days to consider. The range goes from [date - days -1, date]",
metavar="DAYS", default='60')
warnings.simplefilter("ignore", SparseEfficiencyWarning)
(options, _) = parser.parse_args()
if not (options.input_dir and options.iteractions and options.type):
parser.print_usage()
exit(1)
if options.date:
reference = datetime.datetime.strptime(options.date, '%Y%m%d')
else:
reference = datetime.datetime.today().replace(hour=0, minute=0, second=0, microsecond=0)
start = reference - datetime.timedelta(days=int(options.days))
data_iterator = read_data(options.input_dir, start, reference)
(user_relevance, content_relevance, content_to_user,
user_photo, content_to_date, content_to_entities) = compute_relevance(data_iterator,
int(options.iteractions), float(options.damping_factor), options.user, reference, start)
user_relevance_table = {}
outfile = '{0}/{1}_{2}'.format(options.output, options.type[0], reference.strftime('%Y%m%d'))
with open(outfile, 'w') as output_file:
for user in xrange(0, len(user_relevance)):
user_relevance_table[user_relevance[user][0]] = user_relevance[user][1]
if options.type == "USER":
if options.user != user_relevance[user][0]:
u = user_relevance[user][0]
try:
print >> output_file, (u'{0}\t{1}\t{2}'.format(
user_relevance[user][0],
user_relevance[user][1] * 1000000,
user_photo[u])).encode('utf8')
except:
print "Erro de encoding"
if options.type == "CONTENT":
for content in xrange(0, len(content_relevance)):
user = content_to_user[content_relevance[content][0]]
data = content_to_date[content_relevance[content][0]]
entities = content_to_entities[content_relevance[content][0]]
print >> output_file, '{0}\t{1}\t{2}\t{3}\t{4}\thttp://twitter.com/{3}/status/{0}\t{5}'.format(
content_relevance[content][0], content_relevance[content][1] * 1000000,
user_photo[user], user, data, entities)
logger.info('All done. Thanks for flying with us. Check output file %s', outfile)
if __name__ == '__main__':
parser = OptionParser()
parser.add_option("-i", "--input", dest="input", help=u"Input file", metavar="INPUT")
parser.add_option("-n", "--iteractions", dest="iteractions", help=u"Number of iteractions", metavar="ITERACTIONS")
parser.add_option("-d", "--damping-factor", dest="damping_factor", help=u"Damping factor", metavar="DAMPING", default='0.85')
parser.add_option("-o", "--output", dest="output", help=u"Output file", metavar="OUPUT")
parser.add_option("-t", "--type", dest="type", help=u"Type (USER or CONTENT)", metavar="TYPE")
parser.add_option("-u", "--user", dest="user", help=u"User (screen name in Twitter)", metavar="USER", default='')
(options, args) = parser.parse_args()
if not (options.input and options.iteractions and options.output and options.type):
parser.print_usage()
exit(1)
(user_relevance, content_relevance, content_to_user, user_photo) = compute_relevance(options.input,
int(options.iteractions), float(options.damping_factor), options.user)
#(user_num_retweets,user_average_num_retweets,user_num_retweeters,user_sum_num_retweets_retweeter) = compute_user_statistics(input_file_name)
user_relevance_table = {}
with open(options.output, 'w') as output_file:
#output_file.write("user,influence,retweets,avg retweets,retweeters,sum retweeters\n")
for user in xrange(0, len(user_relevance)):
user_relevance_table[user_relevance[user][0]] = user_relevance[user][1]
if options.type == "USER":
if options.user != user_relevance[user][0]:
print >> output_file, '{0},{1}'.format(user_relevance[user][0], user_relevance[user][1])
