gistfile1.py

# -*- coding: utf-8 -*-
from requests_oauthlib import OAuth1
import requests
import json
import re
from datetime import datetime,timedelta

import secret

def streaming(minutes):
    consumerKey=secret.twDict['consumer_key']
    consumerSecret=secret.twDict['consumer_secret']
    accessToken=secret.twDict['access_token_key']
    accessSecret=secret.twDict['access_token_secret']

    url='https://stream.twitter.com/1.1/statuses/sample.json'

    auth=OAuth1(consumerKey,consumerSecret,accessToken,accessSecret)
    res=requests.get(url,auth=auth,stream=True)

    start=datetime.now()
    range_time=timedelta(minutes=minutes)

    cnt = {}
    cnt['w'] = [0 for i in range(4)]
    cnt['W'] = [0 for i in range(4)]
    cnt[u'ｗ'] = [0 for i in range(4)]
    
    for line in res.iter_lines():
        if line:
            data=json.loads(line)
            try:
                if data[u'lang']=='ja':
                    if data[u'text'].find('http')==-1:
                        res = searchWWW(data['text'].encode('utf-8'))
                        print res
                        for k,lst in cnt.items():
                            for i in range(4):
                                lst[i] += res[k][i]
            except:
                pass
        end=datetime.now()
        if end>start+range_time:
            break
    return [start.strftime("%Y/%m/%d %H:%M:%S"),(start+range_time).strftime("%Y/%m/%d %H:%M:%S"),cnt]

def searchWWW(s):
    ss = {}
    ss['w'] = [u'w',u'ww',u'www',u'www(w+)']
    ss['W'] = [u'W',u'WW',u'WWW',u'WWW(W+)']
    ss[u'ｗ'] = [u'ｗ',u'ｗｗ',u'ｗｗｗ',u'ｗｗｗ(ｗ+)']
    
    result = {}
    result['w'] = [0]*4
    result['W'] = [0]*4
    result[u'ｗ'] = [0]*4


    for k,lst in ss.items():
        for i in range(4):
            searchstring = u'(^|[^(\w|ｗ)])(' + lst[i] + u')($|[^(\w|ｗ)])'
            result[k][i] = len(re.findall(searchstring,s))
    return result

def main():
    result = streaming(1)
    print result

if __name__=='__main__':
    main()