基础知识

yield和yield from的使用

yield可以理解为产生一个值，在函数里使用，会把该函数变成generator，它和return的区别在于，yield在返回值之后，代码段不会直接结束，而是中断，同时下一次调用（send()和next()方法）的时候会从这次中断的地方继续执行，即保留了上下文。

def test():
    for i in range(3):
        print(f'yield start')
        yield i
        print('end of yield')

a = test()
print(next(a))
print(next(a))
print(next(a))
b = test()
print(next(b))

# Output:
yield start
0
end of yield
yield start
1
end of yield
yield start
2
yield start
0

当next(a)超出生成器范围时，会报StopIteration异常。

def test():
    for i in range(3):
        yield i

a = test()
print(next(a))
print(next(a))
print(next(a))

# Output:
0
1
StopIteration异常

可以用for来避免该异常

def test():
    for i in range(3):
        yield i

a = test()
for i in a:
    print(i)

# Output:
0
1

可以使用send()方法向generator中传值

def test():
    a = 0
    for i in range(3):
        b = yield a
        a = b + 1

gen = test()
print(gen)
print(next(gen)) # gen.send(None)等价于next(gen)，这里必须先要第一次喂值
print(gen.send(1))
print(gen.send(10))

# Output:
0
2
11

查看生成器的状态

def simple_coro(a):
    print('-> Started: a=',a)
    b = yield a # 相当于是先返回a，暂停，直到next方法调用后，b赋值为send发送过来的数据，再执行到下一个yield暂停
    print('-> Received: b =',b)
    c = yield a+b # 先返回a+b，暂停，直到next方法调用后，c赋值为send发送过来的数据，再执行到下一个yield暂停
    print('-> Received: c=',c)

my_coro2=simple_coro(14) # 实例化一个协程对象

from inspect import getgeneratorstate # 得到状态的函数
print(getgeneratorstate(my_coro2))
next(my_coro2) 
print(getgeneratorstate(my_coro2))
my_coro2.send(28) 
my_coro2.send(99)   
# Outputs:
GEN_CREATED
-> Started: a= 14
GEN_SUSPENDED
-> Received: b = 28
-> Received: c= 99