Official definition: Decorator is a very famous design pattern, which is often used in scenarios with cross-cutting requirements. The more classic ones include inserting logs, performance testing, transaction processing, etc. Decorators are an excellent design to solve this kind of problem. With decorators, we can extract the same code in a large number of functions that has nothing to do with the function itself and continue to reuse it. In a nutshell, the purpose of a decorator is to add additional functionality to an existing object.
Python includes a total of three built-in decorators:
① staticmethod
② classmethod
③ property
Usually, when we access attributes and assign values to attributes, we deal with classes and instances __dict__; but if we want to standardize attribute access, there are two ways available: ① Data descriptor, ②. property() property function.
However, we know that descriptors are relatively complex and difficult to use for novices, so you might as well try property(). Compared to the large process of descriptors, property is equivalent to a thread.
property(fget=None, fset=None, fdel=None, doc=None)
Suppose there is a private variable __x in calss Normal, as shown in the following code:
#code 1 class Normal: def __init__(self): self.__x = None def getx(self): return self.__x def setx(self, value): self.__x = value def delx(self): del self.__x tN = Normal() print(tN.__count)
Output result (an error is reported)
Traceback (most recent call last): File "C:/Users/Administrator/AppData/Local/Programs/Python/Python35/property.py", line 15, in <module> print(tN.__count) AttributeError: 'Normal' object has no attribute '__count'
Why is an error reported? Because the attribute __x of instance tN is a private attribute and cannot be accessed directly, we can only call the internally defined method;
tN = Normal() tN.setx(10) print(tN.getx())
Output result:
6 10
Using the internal method, you can easily Get the private attribute value of the instance or class;
However, if I want to change the setx method name of class Normal to something else (such as Normal_setx), this function is used in many external places, do I need it? Find the calling location of the method one by one, and then change it one by one?
C language may be able, but Python, a high-level language, how can it not solve such a problem?
So, how to solve the above problems?
There are actually two methods.
class Normal: def __init__(self): self.__x = None def getx(self): print('getx(): self.__x=', self.__x) return self.__x def setx(self, value): self.__x = value print('setx()') def delx(self): print('delx()') del self.__x y = property(getx, setx, delx, "I'm a property") tN=Normal() tN.y=10 tN.y del tN.y #输出结果: setx() getx(): self.__x= 10 delx()
Directly operate the method as a property, which is very convenient
class Normal: def __init__(self): self.__x = None @property def xx(self): print('getx(): self.__x=', self.__x) return self.__x @xx.setter def xx(self, value): self.__x = value print('setx()') @xx.deleter def xx(self): print('delx()') del self.__x tN=Normal() tN.xx=10 tN.xx del tN.xx #输出结果信息: setx() getx(): self.__x= 10 delx()
outputs the same result as method 1, which proves that both methods are feasible (note: the first one must be @property (replacing getter, otherwise an error will be reported)).
The above is the detailed content of Detailed explanation of the use of Python's property() decorator. For more information, please follow other related articles on the PHP Chinese website!