目录python
def f1(): x = 10 def f2(): print(x) # 10 x = 1000 f1() # 10 print(x) # 1000
def f1(): x = 10 def f2(): print(x) # 10 return f2 f = f1() # f2 f() # f2()
def f1(x): def f2(): print(x) # 10 return f2 f3 = f1(10) # f2 f3() # f2() # 10 f3() # 10 f3() # 10 f4 = f1(5) f4() # 5 f4() # 5
def f1(x): def f2(): print(x) # 10 return f2 f2 = f1() f2() # f2()
def login_deco(func): def wrapper(*args,**kwargs): login_judge = login() if login_judge: res = func(*args,**kwargs) return res return wrapper @login_deco def shopping(): pass # shopping = deco(shopping) # shopping()
def sanceng(x,y): def login_deco(func): print(x,y) def wrapper(*args,**kwargs): login_judge = login() if login_judge: res = func(*args,**kwargs) return res return wrapper return login_deco @sanceng(10,20) def shopping(): pass day20 # shopping = login_deco(shopping) # shopping()
自定义的迭代器,函数内部使用yield关键,有yield关键字的函数只要调用,这个调用后的函数就是生成器编程
def f1(): yield 1 g = f1() # 变成生成器 for i in g: print(i) # 1
递归本质上就是函数调用函数自己,必须得有结束条件,而且在递归的过程当中,问题的规模必须都不断缩小闭包
def find_num(num,lis): if len(lis) == 1 and lis[0] != num: print('没找到') return mid_ind = int(len(lis) / 2) # 中间索引 mid_num = lis[mid_ind] # 中间值 if num < mid_num: lis = lis[:mid_ind] find_num(num,lis) elif num > mid_num: lis = lis[mid_ind + 1:] find_num(num, lis) else: print('find') lis = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19] find_num(20,lis)
lamdbda 参数 : 逻辑代码
app
max(dic,key=lambda name: dic[name]) max(dic) max(lis/se/tup)
相似于工厂的流水线,机械式的一步一步完成一个项目,把完成步骤具体细分,这样步骤与步骤之间互不干涉函数
缺点:扩展性差,只要有一个步骤断了,项目就崩溃了
优势:清晰优雅code