I have an underlying class which I want to place in some code. I only want it to be instantiated or started once for a given app although it might be called many times.. The problem with the code below is that LowClass is started over and over again. I only want it to start once per test..
import logging
class LowClass:
active = False
def __init__(self):
self.log = logging.getLogger()
self.log.debug("Init %s" % self.__class__.__name__)
if self.active:
return
else:
self.active = True
self.log.debug("Now active!")
class A:
def __init__(self):
self.log = logging.getLogger()
self.log.debug("Init %s" % self.__class__.__name__)
self.lowclass = LowClass()
class B:
def __init__(self):
self.log = logging.getLogger()
self.log.debug("Init %s" % self.__class__.__name__)
self.lowclass = LowClass()
class C:
def __init__(开发者_开发问答self):
self.log = logging.getLogger()
self.log.debug("Init %s" % self.__class__.__name__)
self.a = A()
self.b = B()
class ATests(unittest.TestCase):
def setUp(self):
pass
def testOne(self):
a = A()
b = B()
def testTwo(self):
c = C()
Thanks for pointing out my problem!!
See singleton in python.
What you need is to implement the Singleton Design pattern in python
I've found this implementation I hope it helps
Look up "singleton pattern". It looks like Python you are dealing with, so you might have to search a bit harder. I now there are plenty of examples in C++.
When you import a module in python, it is natural singleton. For example, in util.py, you have:
class MyClass(object):
......
g_my_obj = MyClass()
in a.py, you import g_my_obj:
from util import g_my_obj
in b.py, you import g_my_obj again:
from util import g_my_obj
Safely, there is only one g_my_obj created no matter how many times you import it.
I believe that the problem with the code shown is that you set self.active=True
. This will create an attribute for the instance and set that to True
, whereas I believe that you want to set the active of the class itself to True
. So, instead of self.active=True
, try self.__class__.active=True
.
精彩评论