mongo_cache.py创建

import pickle
import zlib
from datetime import datetime,timedelta

import requests
from pymongo import MongoClient
from bson.binary import Binary

class MongoCache(object):
“”"
数据库缓存
“”"
def init(self,client=None,expires=timedelta(days=30)):
self.client = MongoClient(“localhost”,27017)
self.db = self.client.cache
##加速查找设置索引，设置超时时间,如果达到expi reAfterSeconds设置的超时时间，mongodb会自动删除超时数据
self.db.webpage.create_index(‘timestamp’,expireAfterSeconds=expires.total_seconds())

def __setitem__(self, key, value):
    # 压缩数据设置时间戳
    record = {"result":Binary(zlib.compress(pickle.dumps(value))),"timestamp":datetime.utcnow()}
    ##使用update的upsert (如果不存在执行insert,存在update)参数迸行插入更新操作，$set内置函数表示覆盖原始数据
    self.db.webpage.update({"_id":key},{'$set':record},upsert=True)

def __getitem__(self, item):
    #根据_id以item作为关键字，查找相关网页
    record = self.db.webpage.find_one({"_id":item})
    if record:
        #如果存在进行解压缩反序列化
        return pickle.loads(zlib.decompress(record["result"]))
    else:
        raise KeyError(item + "does not exist")#找不到抛出异常

def __contains__(self, item):
    try:
        self[item]#执行__getitem__方法
    except KeyError:
        return False#捕获到keyerror异常
    else:
        return True#找到相应数据
def clear(self):
    self.db.webpage.drop()