Python从MongoDB中按天读取数据并格式化日志

ykwll

#$cat SpeechMongoHandle.py　　

from pymongo import Connection　　

import time　　

import datetime　　

　　

# CTRL_A='\x01'　　
#
CTRL_B='\x02'　　

　　
CTRL_A='--'
　　

　　
def getEveryDay(begin_date,end_date):
　　date_list = []
　　begin_date = datetime.datetime.strptime(begin_date, "%Y-%m-%d")
　　end_date = datetime.datetime.strptime(end_date,"%Y-%m-%d")
　　while begin_date < end_date:
　　# date_str = begin_date.strftime("%Y-%m-%d")
　　mid_date = begin_date + datetime.timedelta(days=1)
　　tup = (str(begin_date),str(mid_date))
　　date_list.append(tup)
　　begin_date = mid_date
　　return date_list
　　
days = getEveryDay('2017-01-01','2017-01-02')
　　

　　

　　
# MongoDB Connect
　　
client = Connection('syslog-1',27017)
　　
db_name = 'service'
　　
db = client[db_name]
　　
collection = db.speech
　　

　　
for day in days:
　　tmpArray0 = time.strptime(str(day[0]), "%Y-%m-%d %H:%M:%S")
　　tmpArray1 = time.strptime(str(day[1]), "%Y-%m-%d %H:%M:%S")
　　timestamp0 = int(time.mktime(tmpArray0)*1000)
　　timestamp1 = int(time.mktime(tmpArray1)*1000)
　　print day
　　print (timestamp0,timestamp1)
　　

　　results = collection.find({"time":{'$gte':timestamp0,'$lt':timestamp1}})
　　

　　for res in results:
　　keys = res.keys()
　　if "time" in keys:
　　unixtimestamp = str(int(res["time"]))
　　timetemp = time.localtime(int(unixtimestamp)/1000)
　　thedate = time.strftime("%Y-%m-%d %H:%M:%S", timetemp)
　　if "id" in keys:
　　sn = res["id"]
　　if "asr" in keys:
　　asr = res["asr"]
　　if "nlp" in keys:
　　nlp = res["nlp"]
　　if "domain" in keys:
　　domain = res["domain"]
　　if "intent" in keys:
　　intent = res["intent"]
　　

　　print thedate + CTRL_A + sn + CTRL_A + asr + CTRL_A + nlp + CTRL_A + domain + CTRL_A + intent