Object Cacher

Simple cacher for your objects. Store copy of objects in memory (or pickle in file) and returns copy of that objects. Function arguments it is a cache key.

>>> from object_cacher import ObjectCacher
>>> @ObjectCacher(timeout=5)
... def test(*args):
...     print ('Real call')
...     return args
...
>>> test(1,2,3)
Real call
(1, 2, 3)

>>> test(1,2,3)
(1, 2, 3)

>>> test(1,2,3,4)
Real call
(1, 2, 3, 4)

>>> test(1,2,3,4)
... (1, 2, 3, 4)

Makes cache for results of hard functions or methods. For example you have remote RESTful api with a lot of dictionaries. You may cache it:

>>> from urllib import urlopen
>>> from object_cacher import ObjectCacher
>>> @ObjectCacher(timeout=60)
... def get_api():
...     print "This real call"
...     return urlopen('https://api.github.com/').read()
...
>>> get_api()
This real call
'{"current_user_url":"https://api.github.com/user", ...'
>>> get_api()
'{"current_user_url":"https://api.github.com/user", ...'

As result you made http request only once.

For methods you may use it like this:

>>> from urllib import urlopen
>>> from object_cacher import ObjectCacher
>>> class API(object):
...     @ObjectCacher(timeout=60, ignore_self=True)
...     def get_methods(self):
...         print "Real call"
...         return urlopen('https://api.github.com/').read()
...
>>> a = API()
>>> a.get_methods()
Real call
'{"current_user_url":"https://api.github.com/user", ...'
>>> b = API()
>>> b.get_methods()
'{"current_user_url":"https://api.github.com/user", ...'

If ignore_self parameter is set, cache will be shared by all instances. Otherwise cache for instances will be split.

Also you may use persistent cache. The “ObjectPersistentCacher” class-decorator makes file-based pickle-serialized cache storage. When you want to keep cache after rerun you must determine cache id:

>>> from urllib import urlopen
>>> from object_cacher import ObjectCacher
>>> class API(object):
...     @ObjectPersistentCacher(timeout=60, ignore_self=True, oid='com.github.api.listofmethods')
...     def get_methods(self):
...         print "Real call"
...         return urlopen('https://api.github.com/').read()
...
>>> a = API()
>>> a.get_methods()
Real call
'{"current_user_url":"https://api.github.com/user", ...'
>>> b = API()
>>> b.get_methods()
'{"current_user_url":"https://api.github.com/user", ...'

That is keep cache after rerun.

You may change cache dir for ObjectPersistentCacher via changing ‘CACHE_DIR’ class-property.

>>> ObjectPersistentCacher.CACHE_DIR = '/var/tmp/my_cache'