装饰器:原函数基础 ------> 动态添加功能
一、不带参数的decorator
#decorator封装
def log(func):
def wrapper(*args, **kw):
print 'call %s():' %func.__name__
return func(*args, **kw)
return wrapper
#此函数一定要放在 def log(func): 之后
@log
def now():
print "haha"
now()
log(now())
print "now_name:%s"%now.__name__
简单的图解:(待以后深入理解后再详细补)
二、带参数的docurator
#!/usr/bin/python
#docurator封装
def log(text):
def decorator(func):
def wrapper(*args, **kw):
print('%s %s():' % (text, func.__name__))
return func(*args, **kw)
return wrapper
return decorator
@log('execute')
def now():
print "haha"
print "now_name:%s"%now.__name__
now()
print log('execute')
print log(now)
print log('execute')(now)
三、完整的decorator的写法:Python内置的functools.wraps
import functools #导入functools模块
def log(text):
def decorator(func):
@functools.wraps(func)
def wrapper(*args, **kw):
print('%s %s():' % (text, func.__name__))
return func(*args, **kw)
return wrapper
return decorator
@log('execute')
def now():
print "haha"
print "now_name:%s"%now.__name__
now()
print log('execute')
print log(now)
print log('execute')(now)
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
日期:2018/8/18
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
补充:
多个装饰器的执行顺序是无序的,但执行过程是有序的。
装饰器执行过程是:遇到外接函数就跳转到另一个装饰器中执行,若再另一个装饰器中遇到同样的外接函数,则继续执行
证明1:多个装饰器的执行顺序是无序的
#!/usr/bin/python # -*- coding: utf-8 -*- import time def extend(func): def wrapper(*args, **kwargs): print "Begin AAAAAA" func(*args, **kwargs) print("A:%s"%(func.__name__)) print(reduce(lambda x,y:x+y, args), kwargs) print "End AAAAA" return wrapper def extend1(func): def wrapper(*args, **kwargs): print "Begin BBBBB" func(*args, **kwargs) print("B:%s"%(func.__name__)) print(map(lambda x:x*2, args), kwargs) print "End BBBBB" return wrapper @extend @extend1 def get_time1(*args,**kwargs): begin_time = time.time() for n in range(3,4): time.sleep(1) end_time = time.time() print("run %d "% (end_time - begin_time)) #print(reduce(lambda x,y:x+y, args), kwargs) if __name__ == "__main__": get_time1(1,2,3,4,5,6,ma=1,du=2)
执行结果:先从第一个装饰器标志@extend开始
Begin AAAAAA Begin BBBBB run 1 B:get_time1 ([2, 4, 6, 8, 10, 12], {'du': 2, 'ma': 1}) End BBBBB A:wrapper (21, {'du': 2, 'ma': 1}) End AAAAA
另一个例子:
#!/usr/bin/python # -*- coding: utf-8 -*- def dec1(func): print("1111") def wrapper(*args, **kwargs): print("2222") func() print("3333") return wrapper def dec2(func): print("aaaa") def wrapper(*args, **kwargs): print("bbbb") func() print("cccc") return wrapper @dec1 @dec2 def test(): print("test test") test()
运行结果:从第二个装饰器标志@dec2开始
aaaa 1111 2222 bbbb test test cccc 3333
通过比较,多个解释器的执行顺序是无序的。若大家觉得证明不够严谨,可自己多添几个试试。
证明2:装饰器的执行过程是有序的
图1:
图2:
没看懂?没关系,进行步骤解析:
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
2018/8/20
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
补充:一个讲的很好的BLOG: https://blog.csdn.net/xiangxianghehe/article/details/77170585
免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。