Parse the Twitter stream for #tibet or #freetibet and identify potential spammers based on static traits.

clean_tibet.py

import pycurl
import simplejson as json
import urllib
import time
import datetime
  
STREAM_URL = "https://stream.twitter.com/1/statuses/filter.json"  
  
USER = "YOUR USERNAME"  
PASS = "YOUR PASSWORD"
SPAMMERS = []  
data = { "track":"tibet,freetibet" }  
data = urllib.urlencode(data)

def spam_sent(data):
	do = None

def handle_spam(user):
	if user not in SPAMMERS:
		conn = pycurl.Curl()
		conn.setopt(pycurl.USERPWD, "%s:%s" % (USER, PASS))
		conn.setopt(pycurl.URL, "http://api.twitter.com/1/report_spam.json")
		conn.setopt(pycurl.POSTFIELDS,urllib.urlencode({"screen_name":user}))
		conn.setopt(pycurl.WRITEFUNCTION, spam_sent)
		conn.perform()
		if conn.getinfo(pycurl.HTTP_CODE) == 200:
			SPAMMERS.append(user)
			print "== ABUSE SENT for " + user + " =="

def on_receive(data):  
	try:
		d = json.loads(data)
		followers = d['user']['followers_count']
		birth = time.strftime('%Y-%m-%d', time.strptime(d['user']['created_at'],'%a %b %d %H:%M:%S +0000 %Y'))	
		user = d['user']['screen_name']
		current = datetime.datetime.now()
		birth = datetime.datetime.strptime(birth,"%Y-%m-%d")
		delta = current - birth
		tweet = d['text']
	
		if user not in SPAMMERS:
			if (followers >= 5 or delta.days >= 5) and (len(tweet) > 10):
				tweet_time = time.strftime('%Y-%m-%d %H:%M:%S', time.strptime(d['created_at'],'%a %b %d %H:%M:%S +0000 %Y'))
				print user + ": " + tweet + " - " + tweet_time
			else:
				print "!POTENTIAL SPAMMER - " + user + " flagged: FOLLOWERS - " + str(followers) + " ALIVE - " + str(delta.days) + " days TWEET LENGTH - " + str(len(tweet))
				handle_spam(user) 
	except:
		print "!TWITTER ERROR!"

conn = pycurl.Curl()  
conn.setopt(pycurl.USERPWD, "%s:%s" % (USER, PASS))  
conn.setopt(pycurl.URL, STREAM_URL)  
conn.setopt(pycurl.WRITEFUNCTION, on_receive)  
conn.setopt(pycurl.POSTFIELDS,data)
conn.perform()