Home > Backend Development > Python Tutorial > Introduction to the use of Python black magic descriptors

Introduction to the use of Python black magic descriptors

高洛峰
Release: 2017-03-17 17:36:37
Original
1176 people have browsed it

Introduction

Descriptors (descriptors) are a profound but important black magic in the Python language. It is widely used in the kernel of the Python language. , mastering descriptors will add an extra skill to the Pythonprogrammer's toolbox. In this article, I will describe the definition of descriptors and some common scenarios, and at the end of the article I will add getattr, getattribute, getitem, which also involve attributes. Access the Magic Method.

Definition of descriptor

descrget(self, obj, objtype=None) --> value
descr.set(self, obj, value) --> None
descr.delete(self, obj) --> None
Copy after login

As long as one<a href="//m.sbmmt.com/wiki/60.html" target="_blank">object</a> attribute(Objectattribute) defines one of the above three methods Any one of them, then this class can be called a descriptor class.

Descriptor Basics

In the following example we create a RevealAcess class and implement the get method. Now this class can be called is a descriptor class.

class RevealAccess(object):
    def get(self, obj, objtype):
        print(&#39;self in RevealAccess: {}&#39;.format(self))
        print(&#39;self: {}\nobj: {}\nobjtype: {}&#39;.format(self, obj, objtype))
class MyClass(object):
    x = RevealAccess()
    def test(self):
        print(&#39;self in MyClass: {}&#39;.format(self))
Copy after login

EX1 instance attributes

Next let’s take a look at the meaning of each parameter of the get method. In the following example, self is the instance x of the RevealAccess class, obj is the instance m of the MyClass class, objtype as the name suggests is the MyClass class itself. As can be seen from the output statement, m.xaccess descriptor x will call the get method.

>>> m = MyClass()
>>> m.test()
self in MyClass: <main.MyClass object at 0x7f19d4e42160>
>>> m.x
self in RevealAccess: <main.RevealAccess object at 0x7f19d4e420f0>
self: <main.RevealAccess object at 0x7f19d4e420f0>
obj: <main.MyClass object at 0x7f19d4e42160>
objtype: <class &#39;main.MyClass&#39;>
Copy after login

EX2 class attribute

If the attribute x is directly accessed through the class, then obj is directly None, which is still It's easier to understand, because there is no instance of MyClass.

>>> MyClass.x
self in RevealAccess: <main.RevealAccess object at 0x7f53651070f0>
self: <main.RevealAccess object at 0x7f53651070f0>
obj: None
objtype: <class &#39;main.MyClass&#39;>
Copy after login

The principle of descriptor

Descriptor trigger

In the above example, we enumerated the usage of descriptors from the perspective of instance attributes and class attributes. Below we Let’s carefully analyze the internal principle:

  • If you access the instance attribute, the getattribute method of the base class object is actually called. In this method, obj.d is translated into type(obj).dict['d'].get(obj, type(obj)).

  • If you access the class attribute , it is equivalent to calling the getattribute method of the metaclass type, which translates cls.d into cls. dict['d'].get(None, cls), here the obj of get() is None, because there is no instance.

Let’s briefly talk about the getattribute magic method. This method will be called unconditionally when we access the attributes of an object. The detailed details are such as getattr I will make an additional supplement at the end of the article about the difference between , getitem, but we will not delve into it for now.

DescriptorPriority

First of all, descriptors are divided into two types:

  • If an object defines get at the same time () and set() methods, this descriptor is called data descriptor.

  • If an object only defines the get() method, this descriptor is called non-data descriptor.

There are four situations when we access properties:

  • data descriptor

  • instance dict

  • non-data descriptor

  • getattr()

Their priorities The size is:

data descriptor > instance dict > non-data descriptor > getattr()
Copy after login

What does this mean? That is to say, if the data descriptor->d and instance attribute->d with the same name appear in the instance object obj, obj.d will pair with the attribute dWhen accessing, since the data descriptor has a higher priority, Python will call type(obj).dict['d'].get(obj, type(obj)) instead of calling obj.dict['d']. But if the descriptor is a non-data descriptor, Python will call obj.dict['d'].

Property

Defining a descriptor class every time a descriptor is used seems very cumbersome. Python provides a concise way to add data descriptors to properties.

property(fget=None, fset=None, fdel=None, doc=None) -> property attribute
Copy after login

fget, fset and fdel are the getter, setter and deleter methods of the class respectively. We use the following example to illustrate how to use Property:

class Account(object):
    def init(self):
        self._acct_num = None
    def get_acct_num(self):
        return self._acct_num
    def set_acct_num(self, value):
        self._acct_num = value
    def del_acct_num(self):
        del self._acct_num
    acct_num = property(get_acct_num, set_acct_num, del_acct_num, &#39;_acct_num property.&#39;)
Copy after login

If acct is an instance of Account, acct.acct_num will call the getter, acct.acct_num = value will call the setter, and del acct_num.acct_num will call deleter.

>>> acct = Account()
>>> acct.acct_num = 1000
>>> acct.acct_num
1000
Copy after login

Python also provides the @property decorator, which can be used to create properties for simple application scenarios. A property object has getter, setter and delete decorator methods, which can be used to create a copy of the property through the corresponding accessor function of the decorated function.

class Account(object):
    def init(self):
        self._acct_num = None
    @property
     # the _acct_num property. the decorator creates a read-only property
    def acct_num(self):
        return self._acct_num
    @acct_num.setter
    # the _acct_num property setter makes the property writeable
    def set_acct_num(self, value):
        self._acct_num = value
    @acct_num.deleter
    def del_acct_num(self):
        del self._acct_num
Copy after login

If you want the property to be read-only, just remove the setter method.

在运行时创建描述符

我们可以在运行时添加property属性:

class Person(object):
    def addProperty(self, attribute):
        # create local setter and getter with a particular attribute name
        getter = lambda self: self._getProperty(attribute)
        setter = lambda self, value: self._setProperty(attribute, value)
        # construct property attribute and add it to the class
        setattr(self.class, attribute, property(fget=getter, \
                                                    fset=setter, \
                                                    doc="Auto-generated method"))
    def _setProperty(self, attribute, value):
        print("Setting: {} = {}".format(attribute, value))
        setattr(self, &#39;_&#39; + attribute, value.title())
    def _getProperty(self, attribute):
        print("Getting: {}".format(attribute))
        return getattr(self, &#39;_&#39; + attribute)
Copy after login
>>> user = Person()
>>> user.addProperty(&#39;name&#39;)
>>> user.addProperty(&#39;phone&#39;)
>>> user.name = &#39;john smith&#39;
Setting: name = john smith
>>> user.phone = &#39;12345&#39;
Setting: phone = 12345
>>> user.name
Getting: name
&#39;John Smith&#39;
>>> user.dict
{&#39;_phone&#39;: &#39;12345&#39;, &#39;_name&#39;: &#39;John Smith&#39;}
Copy after login

静态方法和类方法

我们可以使用描述符来模拟Python中的@<a href="//m.sbmmt.com/wiki/188.html" target="_blank">static</a>method@classmethod的实现。我们首先来浏览一下下面这张表:

TransformationCalled from an ObjectCalled from a Class
functionf(obj, *args)f(*args)
staticmethodf(*args)f(*args)
classmethodf(type(obj), *args)f(klass, *args)

静态方法

对于静态方法fc.fC.f是等价的,都是直接查询object.getattribute(c, ‘f’)或者object.getattribute(C, ’f‘)。静态方法一个明显的特征就是没有self变量

静态方法有什么用呢?假设有一个处理专门数据的容器类,它提供了一些方法来求平均数,中位数等统计数据方式,这些方法都是要依赖于相应的数据的。但是类中可能还有一些方法,并不依赖这些数据,这个时候我们可以将这些方法声明为静态方法,同时这也可以提高代码的可读性。

使用非数据描述符来模拟一下静态方法的实现:

class StaticMethod(object):
    def init(self, f):
        self.f = f
    def get(self, obj, objtype=None):
        return self.f
Copy after login

我们来应用一下:

class MyClass(object):
    @StaticMethod
    def get_x(x):
        return x
print(MyClass.get_x(100))  # output: 100
Copy after login

类方法

Python的@classmethod@staticmethod的用法有些类似,但是还是有些不同,当某些方法只需要得到类的<a href="//m.sbmmt.com/wiki/231.html" target="_blank">引用</a>而不关心类中的相应的数据的时候就需要使用classmethod了。

使用非数据描述符来模拟一下类方法的实现:

class ClassMethod(object):
    def init(self, f):
        self.f = f
    def get(self, obj, klass=None):
        if klass is None:
            klass = type(obj)
        def newfunc(*args):
            return self.f(klass, *args)
        return newfunc
Copy after login

其他的魔术方法

首次接触Python魔术方法的时候,我也被get, getattribute, getattr, getitem之间的区别困扰到了,它们都是和属性访问相关的魔术方法,其中重写getattrgetitem来构造一个自己的集合类非常的常用,下面我们就通过一些例子来看一下它们的应用。

getattr

Python默认访问类/实例的某个属性都是通过getattribute来调用的,getattribute会被无条件调用,没有找到的话就会调用getattr。如果我们要定制某个类,通常情况下我们不应该重写getattribute,而是应该重写getattr,很少看见重写getattribute的情况。

从下面的输出可以看出,当一个属性通过getattribute无法找到的时候会调用getattr

In [1]: class Test(object):
    ...:     def getattribute(self, item):
    ...:         print(&#39;call getattribute&#39;)
    ...:         return super(Test, self).getattribute(item)
    ...:     def getattr(self, item):
    ...:         return &#39;call getattr&#39;
    ...:
In [2]: Test().a
call getattribute
Out[2]: &#39;call getattr&#39;
Copy after login

应用

对于默认的字典,Python只支持以obj[&#39;foo&#39;]形式来访问,不支持obj.foo的形式,我们可以通过重写getattr让字典也支持obj[&#39;foo&#39;]的访问形式,这是一个非常经典常用的用法:

class Storage(dict):
    """
    A Storage object is like a dictionary except `obj.foo` can be used
    in addition to `obj[&#39;foo&#39;]`.
    """
    def getattr(self, key):
        try:
            return self[key]
        except KeyError as k:
            raise AttributeError(k)
    def setattr(self, key, value):
        self[key] = value
    def delattr(self, key):
        try:
            del self[key]
        except KeyError as k:
            raise AttributeError(k)
    def repr(self):
        return &#39;<Storage &#39; + dict.repr(self) + &#39;>&#39;
Copy after login

我们来使用一下我们自定义的加强版字典:

>>> s = Storage(a=1)
>>> s[&#39;a&#39;]
1
>>> s.a
1
>>> s.a = 2
>>> s[&#39;a&#39;]
2
>>> del s.a
>>> s.a
...
AttributeError: &#39;a&#39;
Copy after login

getitem

getitem用于通过下标[]的形式来获取对象中的元素,下面我们通过重写getitem来实现一个自己的list。

class MyList(object):
    def init(self, *args):
        self.numbers = args
    def getitem(self, item):
        return self.numbers[item]
my_list = MyList(1, 2, 3, 4, 6, 5, 3)
print my_list[2]
Copy after login

这个实现非常的简陋,不支持slice和step等功能,请读者自行改进,这里我就不重复了。

应用

下面是参考requests库中对于getitem的一个使用,我们定制了一个忽略属性大小写的字典类。

程序有些复杂,我稍微解释一下:由于这里比较简单,没有使用描述符的需求,所以使用了@property装饰器来代替,lower_keys的功能是将实例字典中的键全部转换成小写并且存储在字典self._lower_keys中。重写了getitem方法,以后我们访问某个属性首先会将键转换为小写的方式,然后并不会直接访问实例字典,而是会访问字典self._lower_keys去查找。赋值/删除操作的时候由于实例字典会进行变更,为了保持self._lower_keys和实例字典同步,首先清除self._lower_keys的内容,以后我们重新查找键的时候再调用getitem的时候会重新新建一个self._lower_keys

class CaseInsensitiveDict(dict):
    @property
    def lower_keys(self):
        if not hasattr(self, &#39;_lower_keys&#39;) or not self._lower_keys:
            self._lower_keys = dict((k.lower(), k) for k in self.keys())
        return self._lower_keys
    def _clear_lower_keys(self):
        if hasattr(self, &#39;_lower_keys&#39;):
            self._lower_keys.clear()
    def contains(self, key):
        return key.lower() in self.lower_keys
    def getitem(self, key):
        if key in self:
            return dict.getitem(self, self.lower_keys[key.lower()])
    def setitem(self, key, value):
        dict.setitem(self, key, value)
        self._clear_lower_keys()
    def delitem(self, key):
        dict.delitem(self, key)
        self._lower_keys.clear()
    def get(self, key, default=None):
        if key in self:
            return self[key]
        else:
            return default
Copy after login

我们来调用一下这个类:

>>> d = CaseInsensitiveDict()
>>> d[&#39;ziwenxie&#39;] = &#39;ziwenxie&#39;
>>> d[&#39;ZiWenXie&#39;] = &#39;ZiWenXie&#39;
>>> print(d)
{&#39;ZiWenXie&#39;: &#39;ziwenxie&#39;, &#39;ziwenxie&#39;: &#39;ziwenxie&#39;}
>>> print(d[&#39;ziwenxie&#39;])
ziwenxie
# d[&#39;ZiWenXie&#39;] => d[&#39;ziwenxie&#39;]
>>> print(d[&#39;ZiWenXie&#39;])
ziwenxi
Copy after login

The above is the detailed content of Introduction to the use of Python black magic descriptors. For more information, please follow other related articles on the PHP Chinese website!

Related labels:
source:php.cn
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template