python-webpy和mongodb实现博客统计

本宝宝从0开始写的博客一直没有博客统计功能，于是计划加上统计功能！！

step:0x001

python使用pymongo登录远程数据库

mongodb默认本地就可以直接访问，设置密码后要这样才能访问

client = MongoClient('host','port')
ret = client['admin'].authenticate('username', 'password')
print ret
return client['dbname']

step:0x010

表设计

#今日ip访问表
# connectStatistics   agent day(like:2015-11-11) ip blogId(if not blogId = -1)

#博客uv,pv历史表
# blogStatistics   day(like:2015-11-11) pv uv

统计思路及代码

先获取connectStatistics表最后插入一条的记录，获取时间(格式：yyyy-mm-dd)，用该时间与系统时间对比，如果不一致，则是昨天的记录，此时统计connectStatistics的pv，uv数据，将输入插入blogStatistics历史访问表，成功后删除访问表的全部数据，开始新一天的博客数据收集统计。

saveDate = time.strftime("%y-%m-%d",time.localtime())

oneRecd =  ut.getMongoDb().connectStatistics.find_one()

if oneRecd:
lastSaveDate = oneRecd['day']

if not saveDate == lastSaveDate:
print 'del old data,the date is:%s'%lastSaveDate

blogStat = {}
blogStat['day'] = lastSaveDate
blogStat['pv'] = but.getPvToday()
blogStat['uv'] = but.getUvToday()

mongodb.blogStatistics.insert(blogStat)
mongodb.connectStatistics.remove({"day":lastSaveDate})

step:0x011

收集今日记录：

在统一访问入口开启ip访问记录线程：

import thread

agent = web.ctx.env['HTTP_USER_AGENT']
remoteAddr = web.ctx.env['REMOTE_ADDR']
params = ut.getInput(web.input())
if params.has_key('id'):
blogId = params['id']
else:
blogId = '-1'
thread.start_new_thread (self.saveIp,(blogId, agent,remoteAddr))

def saveIp (self,blogId,agent,remoteAddr):

if agent==None or len(agent)==0:
return

#文章记录 readCount
blog = mongodb.find_one({'id':blogId})
print "blog:"
print blog
if blog:
#博客阅读数据+1
mongodb.blog.update({"id": blogId}, {"$set": {"readCount": blog['readCount']+1}})
#end

saveDate = time.strftime("%y-%m-%d",time.localtime())

oneRecd =  mongodb.connectStatistics.find_one()

if oneRecd:
lastSaveDate = oneRecd['day']

if not saveDate == lastSaveDate:
print 'del old data,the date is:%s'%lastSaveDate

blogStat = {}
blogStat['day'] = lastSaveDate
# blogStat['pv'] = mongodb.connectStatistics.count()
blogStat['pv'] = but.getPvToday()
blogStat['uv'] = but.getUvToday()

mongodb.blogStatistics.insert(blogStat)
mongodb.connectStatistics.remove({"day":lastSaveDate})

data = {}
data['blogId'] = blogId
data['agent']=agent
data['ip']=remoteAddr
data['day']= saveDate

mongodb.connectStatistics.insert(data)

def getPvToday():
try:
return mongodb.connectStatistics.count()
except Exception, e:
print e
return 0

def getUvToday():
try:
key = ['ip']
cond = None
initial = {'count' : 0}
reduce = "function (obj, prev) { prev.count++; }"
#按ip分组，计算组数
group = list(mongodb.connectStatistics.group(key,cond,initial,reduce))
uv = len(group)
return uv
except Exception, e:
print e
return 0

step:0x100

显示在html

新建工具类返回blog访问记录map：

#获取博客统计
def getWebInfo():
cacheKey = 'blogutil-getwebinfo-key'
ret = lrucache.getinstance().get(cacheKey)
if ret:
print 'use cache:' + cacheKey
return ret

info = {}
info['uv'] = getUvToday()
info['pv'] = getPvToday()
info['readCount'] = int(getBlogPv())

#两分钟的缓存
lrucache.getinstance().set(cacheKey,info,120)

return info

代码获取访问数据，传给templates显示

return ut.getTemplatesWithBase().showcontent(util.getWebInfo(),blogInfo,newLast,monthList,classificationMap)

显示数据

<div class="lm" id="menu_card">
<ul>
网站统计:
<li>
总访问量:$:webInfo['readCount']
</li>

<li>
今日uv:$:webInfo['uv']
</li>

<li>
今日pv:$:webInfo['pv']
</li>
</ul>
</div>