This page is for memcache tips and tricks

 

   An Item of  Interest: There is a posting on web2py forum that there is a missing parameter bug in a call to the memcache API.  http://groups.google.com/group/web2py/browse_thread/thread/fcac81ee1b713dbe  

Sounds like we need a 'lint' tool for python which can catch missing args.  This bug report posting also has an interesting snippet of memcache api code, so that is very instructive to read.

Here is the code snippet from that posting:

class MemcacheClient(Client): 
    def __init__(self, request, servers, debug=0, pickleProtocol=0, 
                 pickler=pickle.Pickler, unpickler=pickle.Unpickler, 
                 pload=None, pid=None): 
        self.request=request 
        Client.__init__(self,servers,debug,pickleProtocol, 
                        pickler,unpickler,pload,pid) 
    def __call__(self,key,f,time_expire=300): 
        key='%s/%s' % (self.request.application,key) 
        dt=time_expire 
        value=None 
        obj=self.get(key) 
        if obj and obj[0]>time.time()-dt: 
            value=obj[1] 
        elif f is None: 
            if obj: self.delete(key) 
        else: 
            value=f() 
            self.set(key,(time.time(),value)) 
        return value 
    def increment(self,key,value=1): 
        key='%s/%s' % (self.request.application,key) 
        obj=self.get(key) 
        if obj: value=obj[1]+value 
        self.set(key,(time.time(),value)) 
        return value

It does not look to be thread safe. If some asynchronous Ajax calls came in from the browser which try to use this cache, there could be trouble.

Here is a posting " Memcache & SQL Query as key " which shows an example of using Memcache 
           http://groups.google.com/group/web2py/browse_thread/thread/81aefcf711b21c6c 

 

 

 

 

From http://groups.google.com/group/web2py/browse_thread/thread/81aefcf711b21c6c 

 

From: NicolasB <nicolas.bn...@gmail.com>

Date: Wed, 6 May 2009 08:45:21 -0700 (PDT)

Local: Wed, May 6 2009 11:45 am 

Subject: Memcache & SQL Query as key

Hello, 

I am using memcache to hold query directly: 

db((db.ad...,cache=(cache.mem,100)) 

 

Web2py creates a key with the query. However memcache does not seem to 

support spaces. 

 

I modified web2py_src_1.61.4\web2py\gluon\contrib\memcache\__init__.py 

to format the key 

 

I don't know if this is the best way to do... 

 

Best regards, 

Nicolas 

 

from gluon.contrib.memcache.memcache import Client 

import time 

 

""" 

example of usage: 

 

cache.memcache=MemcacheClient(request,[127.0.0.1:11211],debug=true) 

""" 

import cPickle as pickle 

class MemcacheClient(Client): 

    def __init__(self, request, servers, debug=0, pickleProtocol=0, 

                 pickler=pickle.Pickler, unpickler=pickle.Unpickler, 

                 pload=None, pid=None): 

        self.request=request 

        Client.__init__(self,servers,debug,pickleProtocol, pickler,unpickler,pload,pid) 

 

    def __call__(self,key,f,time_expire=300): 

        key=self.__keyFormat__(key) 

        dt=time_expire 

        value=None 

        obj=self.get(key) 

        if obj and obj[0]>time.time()-dt: 

            value=obj[1] 

        elif f is None: 

            if obj: self.delete(key) 

        else: 

            value=f() 

            self.set(key,(time.time(),value)) 

        return value 

 

 

    def increment(self,key,value=1): 

        key=self.__keyFormat__(key) 

        obj=self.get(key) 

        if obj: value=obj[1]+value 

        self.set(key,(time.time(),value)) 

        return value 

 

    def __keyFormat__(self,key): 

        return '%s/%s' % (self.request.application,key.replace(' ','_'))