init_redis.py

__author__ = 'multiangle'

from DB_Interface import MySQL_Interface
import redis

class SimpleHash():
    def __init__(self,cap,seed):
        self.cap=cap
        self.seed=seed
    def hash(self,value):
        ret=0
        for i in range(value.__len__()):
            ret+=self.seed*ret+ord(value[i])
        return ((self.cap-1) & ret)

class BloomFilter():
    def __init__(self):
        self.bit_size=1<<15
        self.seeds=[5,7,11,13,31,37,61]
        self.r=redis.StrictRedis(host='127.0.0.1',port=6379,db=0)
        self.hashFunc=[]
        for i in range(self.seeds.__len__()):
            self.hashFunc.append(SimpleHash(self.bit_size,self.seeds[i]))

    def isContains(self,str_input,name):
        if str_input==None:
            return False
        if str_input.__len__()==0:
            return False
        ret=True
        for f in self.hashFunc:
            loc=f.hash(str_input)
            ret=ret & self.r.getbit(name,loc)
        return ret

    def insert(self,str_input,name):
        for f in self.hashFunc:
            loc=f.hash(str_input)
            self.r.setbit(name,loc,1)

dbi=MySQL_Interface(dbname='microblog_spider')
r=redis.StrictRedis(host='127.0.0.1',port=6379,db=0)
query='select uid from user_info_table ;'
uid=dbi.select_asQuery(query)
uid=[x[0] for x in uid]
bf=BloomFilter()
for id in uid:
    bf.insert(id,'user_info_table')