装饰器

函数对修改封闭，对扩展开放
例子：
1.对函数的执行时间进行检测
import time
def foo():
    start = time.time()
    print(‘foo....‘)
    time.sleep(2)
    end = time.time()
    print(‘spend %s‘ % (end - start))
def test():
    start = time.time()
    print(‘test....‘)
    time.sleep(2)
    end = time.time()
    print(‘spend %s‘ % (end - start))
foo()
test()
问题：一旦修改，所有函数都会改变，并且对更多的函数进行同样操作的话，就会使代码出现大量的重复

2.改进方法,定义show_time函数
def foo():
    print(‘foo....‘)
    time.sleep(2)

def show_time(f):
    start = time.time()
    f()
    end = time.time()
    print(‘spend %s‘ % (end - start))
show_time(foo)
问题：该方法改变了原生函数的调用方式

3.引入装饰器，装饰器就是为之前的函数添加新的功能
def foo():
    print(‘foo....‘)
    time.sleep(2)

def test():
    print(‘test....‘)
    time.sleep(2)

def show_time(f):
    def inner():
        start = time.time()
        f()
        end = time.time()
        print(‘spend %s‘ % (end - start))
    return inner

foo = show_time(foo)
foo()
test = show_time(test)
test()

4.python的优雅写法
def show_time(f):
    def inner():
        start = time.time()
        f()
        end = time.time()
        print(‘spend %s‘ % (end - start))
    return inner

@show_time # test = show_time(test)
def test():
    print(‘test....‘)
    time.sleep(2)
test() #test执行部分从test到了show_time

5.再次优化
def show_time(f):
    def inner(*x, **y):
        start = time.time()
        f(*x, **y)
        end = time.time()
        print(‘spend %s‘ % (end - start))
    return inner

@show_time
def add(*a, **b):
    print(sum(a),sum(b[i] for i in b))
    time.sleep(1)
add(1, 2, 5, 6, x=5, y=8)

6.装饰器加参数，直接加入无法传参
def logger(flag=‘‘):
    def show_time(f):
        def inner(*x, **y):
            start = time.time()
            f(*x, **y)
            end = time.time()
            print(‘spend %s‘ % (end - start))
            if flag==‘true‘:
                print(‘日志记录‘)
        return inner
    return show_time

@logger(‘true‘) # @show_time
def add(*a, **b):
    print(sum(a),sum(b[i] for i in b))
    time.sleep(1)
add(1, 2, 5, 6, x=5, y=8)

装饰器

原文：https://www.cnblogs.com/liqilong/p/9087105.html