先来看看如果遇到一个对象，如何判断其是否是这三种类型：
1 from types import GeneratorType2 from collectiuons import Iterable, Iterator3 4 isinstance( xx, GeneratorType )5 isinstance( xx, Iterable )6 isinstance( xx, Iterator )
生成器对象：
生成器是一个通过yield关键字构建的函数，其返回一个generator对象，同时其又是一个iterator对象，因为其实现了__iter__与next方法
In[4]: def gene(k):... for i in xrange(0, k):... yield i... In[5]: ge = gene(6)In[6]: type(ge)Out[6]: generatorIn[8]: dir(ge)Out[8]: [ ...,'__iter__', 'next', ... ] In[9]: isinstance(ge, collections.Iterator)Out[9]: True
生成器的一个特点是只能用一次
In[11]: for i in ge: ... print i... 012345In[12]: ge.next()Traceback (most recent call last): File "/usr/lib/python2.7/dist-packages/IPython/core/interactiveshell.py", line 2820, in run_code exec code_obj in self.user_global_ns, self.user_ns File "", line 1, in ge.next()StopIteration
迭代器对象：
实现了__iter__与__next__方法的对象。
__iter__返回一个实例作为最终使用的iterator对象--for...in...语句初始化时使用，next方法是后面每次迭代iterator对象时调用的方法
class Counter(object): def __init__(self, low, high): self.current = low self.high = high def __iter__(self): # 'Returns itself as an iterator object' return self def next(self): # 'Returns the next value till current is lower than high' if self.current > self.high: raise StopIteration else: self.current += 1 return self.current - 1
>>> c = Counter(5,10) >>> for i in c: ... print(i, end=' ') ... 5 6 7 8 9 10 >>> for i in c: ... pritn i ... # 无输出 >>> c.next()
Traceback (most recent call last): File "" c.next()StopIteration
可迭代对象：
可迭代对象就是用于for...in...循环的
不同于上面两种，其需要可以多次for...in...使用
通过generator写可迭代对象：
将__iter__写成生成器函数。注意for...in...时初始化时调用__iter__函数使得counter为low, 而后执行的都是yield所影响区域的了【1】，除非下次再执行for...in...语句才会调用到counter初始化那句，这样也就实现了复用。但是有个问题是此时Counter的实例就既不是generator也不是iterator对象了。故其也没法调用next方法
class Counter(object): def __init__(self, low, high): self.low = low self.high = high def __iter__(self): counter = self.low while self.high >= counter: yield counter counter += 1obj = Counter(5, 10)print isinstance(obj, Iterator)print isinstance(obj, GeneratorType)print type(obj)for i in obj: print i obj.next()
False
False<class '__main__.Counter'>5678910Traceback (most recent call last):
File "/home/pd/..." obj.next()AttributeError: 'Counter' object has no attribute 'next'
注释【1】：
In [1]: def gener(k): ...: print "====initial====" ...: for i in range(0, k): ...: print "before yield" ...: yield i ...: print "after yield" ...: print "*****end******" ...: In [2]: g = gener(3)In [3]: for i in g: ...: print g ...: ====initial==== before yieldafter yield before yield after yield before yield after yield *****end******
补充：
关于generator的yield与send：
总的来说：yield返回的是个generator对象，send是向generator输送值
1 In [35]: def gen(): 2 ...: while True: 3 ...: print("before x_yield") 4 ...: x=yield 5 ...: print("after x_yield:", x) 6 ...: yield x*2 7 ...: print("after yield") 8 ...: 9 ...: 10 11 In [36]: g = gen()12 13 In [37]: g.send(10)14 ---------------------------------------------------------------------------15 TypeError Traceback (most recent call last)16 in ()17 ----> 1 g.send(10)18 19 TypeError: can't send non-None value to a just-started generator20 21 In [38]: g.__next__(10)22 ---------------------------------------------------------------------------23 TypeError Traceback (most recent call last)24 in ()25 ----> 1 g.__next__(10)26 27 TypeError: expected 0 arguments, got 128 29 In [39]: g.__next__() 30 before x_yield31 # 由上可知， “x=yield”中的yield也会成为__next__()调用的终点，也就是说如果上面要两个__next__()才能走完。 31 # 通过实验发现，其实x=yield的执行并不是同时的，比如send（10），此时相当于到达了x=yield中的yield，而此时再执行send（11）才会执行x=yield这个赋值操作 31 # 所以输出是22而不是第一次就能传入的20【见43，44的输出】32 33 In [40]: g.send(10) 34 after x_yield: 1035 Out[40]: 20 # 进行到yield x*236 37 In [41]: g.__next__()38 after yield39 before x_yield40 41 In [42]: g.send(10)42 after x_yield: 1043 Out[42]: 2044 45 In [43]: g.send(10) #46 after yield47 before x_yield48 49 In [44]: g.send(11)50 after x_yield: 1151 Out[44]: 22
补充：关于send与__next__()
1 In [71]: def gen(): 2 ...: print("enhen") 3 ...: while True: 4 ...: print("before yield") 5 ...: yield "pd the handsome" 6 ...: print("afer yield") 7 ...: 8 9 In [72]: g=gen()10 11 In [73]: g.send(10)12 ---------------------------------------------------------------------------13 TypeError Traceback (most recent call last)14 in ()15 ----> 1 g.send(10)16 17 TypeError: can't send non-None value to a just-started generator18 19 In [74]: g.__next__()20 enhen21 before yield22 Out[74]: 'pd the handsome' 23 # 由上可见，send(xx)并不能用于初始化调用generator, 但是通过 send(None)其实可以，纯粹的相当于74的调用__next__（）24 # 可见，send方法实际也调用了__next__()从而执行到了下一次的yield 25 In [75]: g.send(10)26 afer yield27 before yield28 Out[75]: 'pd the handsome'29 30 In [76]: g.send(20)31 afer yield32 before yield33 Out[76]: 'pd the handsome'
补充：用send实现协程【】
1 def consumer(): 2 r = '' 3 while True: 4 n = yield r 5 if not n: 6 return 7 print('[CONSUMER] Consuming %s...' % n) 8 r = '200 OK' 9 10 def produce(c):11 c.send(None)12 n = 013 while n < 5:14 n = n + 115 print('[PRODUCER] Producing %s...' % n)16 r = c.send(n) // 执行一次c17 print('[PRODUCER] Consumer return: %s' % r)18 c.close()19 20 c = consumer()21 produce(c)22 23 执行结果：24 25 [PRODUCER] Producing 1...26 [CONSUMER] Consuming 1...27 [PRODUCER] Consumer return: 200 OK28 [PRODUCER] Producing 2...29 [CONSUMER] Consuming 2...30 [PRODUCER] Consumer return: 200 OK31 [PRODUCER] Producing 3...32 [CONSUMER] Consuming 3...33 [PRODUCER] Consumer return: 200 OK34 [PRODUCER] Producing 4...35 [CONSUMER] Consuming 4...36 [PRODUCER] Consumer return: 200 OK37 [PRODUCER] Producing 5...38 [CONSUMER] Consuming 5...39 [PRODUCER] Consumer return: 200 OK 挺叼，传入个generator做操作！
补充：关于yield from :
参考：
一篇关于yield的好文：