导入的和定义的类之间的python区别

Question

我想在实例的某些参数（不一定与示例中的类相同）之间创建依赖关系。我想出了以下代码，直到我决定将它移入模块并通过导入使用它。

class objectD(object): 
    def __init__(self,val1,val2,val3): 
    self.val1 = val1 
    self.val2 = val2 
    self.val3 = val3 

    def __str__(self): 
    return str(str(self.val1)+","+str(self.val2)+","+str(self.val3)) 

    def dependence(self,dependent): 
    print "values val1 and val2 of "+str(self)+" now depend on "+dependent 
    self.val1 = eval(dependent).val1 
    self.val2 = eval(dependent).val2 
    self.dependent = dependent 

    def update(self): 
    self.val1 = eval(self.dependent).val1 
    self.val2 = eval(self.dependent).val2 

#test 
obj1 = objectD(350,4,500) 
print obj1 
obj2 = objectD(230,1,1000) 
print obj2 
obj2.dependence("obj1")#problem with imported class occurs there 
print obj2 
obj1.val1 = 1315 
obj1.val2 = 6464 
print obj1 
obj2.update() 
print obj2 

问题似乎在分配依赖的实例通过使用导入的类的类根本不存在根据python。在我看来，调用方法依赖关系（）对实例继续在模块中，而不是在我的脚本。 有没有办法以某种方式修复导入，或者可能是某种不同的方法来创建依赖关系？

Answer 1

应该不需要eval。为什么不简单地做这个呢？

class objectD(object): 
    def __init__(self,val1,val2,val3): 
    self.val1 = val1 
    self.val2 = val2 
    self.val3 = val3 

    def __str__(self): 
    return str(str(self.val1)+","+str(self.val2)+","+str(self.val3)) 

    def dependence(self,dependent): 
    print "values val1 and val2 of "+str(self)+" now depend on "+dependent 
    self.val1 = dependent.val1 
    self.val2 = dependent.val2 
    self.dependent = dependent 

    def update(self): 
    self.val1 = self.dependent.val1 
    self.val2 = self.dependent.val2 

#test 
obj1 = objectD(350,4,500) 
print obj1 
obj2 = objectD(230,1,1000) 
print obj2 
obj2.dependence(obj1)#problem with imported class no longer occurs there 
print obj2 
obj1.val1 = 1315 
obj1.val2 = 6464 
print obj1 
obj2.update() 
print obj2 

这将如预期，因为dependent仅存储参考obj1，对象本身不在状态。所以致电update()的呼叫实际上会拉动新的值。