为什么在描述符类的__get__中需要“ if instance is None”？

Question

我从有效的Python项目31中获得以下示例：

from weakref import WeakKeyDictionary
class Grade(object):
    def __init__(self):
        self._values = WeakKeyDictionary()
    def __get__(self, instance, instance_type):
        if instance is None: return self
        return self._values.get(instance, 0)

    def __set__(self, instance, value):
        if not (0 <= value <= 100):
            raise ValueError('Grade must be between 0 and 100')
        self._values[instance] = value


# Example 16
class Exam(object):
    math_grade = Grade()
    writing_grade = Grade()
    science_grade = Grade()

first_exam = Exam()
first_exam.writing_grade = 82
second_exam = Exam()
second_exam.writing_grade = 75
print('First ', first_exam.writing_grade, 'is right')
print('Second', second_exam.writing_grade, 'is right')

我想不出任何理由都if instance is None: return self在__get__。一个如何能Exam（或其他潜在的类中使用Grade）的实例是None？

Answer 1

None当访问类的描述符时，Python将传入实例。

通过返回self这种情况，您可以访问类上的描述符对象，而不必绕过协议（通过访问ClassObj.__dict__['name_of_descriptor']）。

>>> class DemoDescriptor:
...     def __get__(self, instance, type_):
...         if instance is None:
...             print('Accessing descriptor on the class')
...             return self
...         print('Accessing descriptor on the instance')
...         return 'Descriptor value for instance {!r}'.format(instance)
... 
>>> class DemoClass(object):
...     foo = DemoDescriptor()
... 
>>> DemoClass.foo  # on the class
Accessing descriptor on the class
<__main__.DemoDescriptor object at 0x1041d3c50>
>>> DemoClass.__dict__['foo']  # bypassing the descriptor protocol
<__main__.DemoDescriptor object at 0x1041d3c50>
>>> DemoClass().foo  # on the instance
Accessing descriptor on the instance
'Descriptor value for instance <__main__.DemoClass object at 0x1041d3438>'

这也是和对象的__get__方法实现的工作方式。functionproperty

针对您的特殊情况下，每个的Exam.math_grade，Exam.writing_grade或Exam.science_grade将调用Grade.__get__，在传递None的实例，并Exam为instance_type。