脚本之家,脚本语言编程技术及教程分享平台!
分类导航

Python|VBS|Ruby|Lua|perl|VBA|Golang|PowerShell|Erlang|autoit|Dos|bat|

服务器之家 - 脚本之家 - Python - 对python中的高效迭代器函数详解

对python中的高效迭代器函数详解

2021-04-09 00:26晓东邪 Python

今天小编就为大家分享一篇对python中的高效迭代器函数详解,具有很好的参考价值,希望对大家有所帮助。一起跟随小编过来看看吧

python中内置的库中有个itertools,可以满足我们在编程中绝大多数需要迭代的场合,当然也可以自己造轮子,但是有现成的好用的轮子不妨也学习一下,看哪个用的顺手~

首先还是要先import一下:

?
1
2
3
#import itertools
from itertools import * #最好使用时用上面那个,不过下面的是为了演示比较
  常用的,所以就直接全部导入了

一.无限迭代器

对python中的高效迭代器函数详解

由于这些都是无限迭代器,因此使用的时候都要设置终止条件,不然会一直运行下去,也就不是我们想要的结果了。

1、count()

可以设置两个参数,第一个参数为起始点,且包含在内,第二个参数为步长,如果不设置第二个参数则默认步长为1

?
1
2
3
for x in count(10,20):
 if x < 200:
 print x
?
1
2
3
4
5
6
7
def count(start=0, step=1):
 # count(10) --> 10 11 12 13 14 ...
 # count(2.5, 0.5) -> 2.5 3.0 3.5 ...
 n = start
 while true:
 yield n
 n += step

2、cycle()

可以设置一个参数,且只接受可以迭代的参数,如列表,元组,字符串。。。,该函数会对可迭代的所有元素进行循环:

?
1
2
3
for i,x in enumerate(cycle('abcd')):
 if i < 5:
 print x
?
1
2
3
4
5
6
7
8
9
def cycle(iterable):
 # cycle('abcd') --> a b c d a b c d a b c d ...
 saved = []
 for element in iterable:
 yield element
 saved.append(element)
 while saved:
 for element in saved:
  yield element

3、repeat()

可以设置两个参数,其中第一个参数要求可迭代,第二个参数为重复次数,第二个参数如不设置则无限循环,一般来说使用时都会设置第二个参数,用来满足预期重复次数后终止:

?
1
2
3
#注意如果不设置第二个参数notebook运行可能会宕机
for x in repeat(['a','b','c'],10):
 print x

二.有限迭代器

对python中的高效迭代器函数详解

1、chain()

可以接受不定个数个可迭代参数,不要求可迭代参数类型相同,会返回一个列表,这个类似于list的extend,不过不同点是list的extend是对原变量进行改变不返回,而chain则是就地改变并返回:

?
1
2
3
4
5
list(chain(range(4),range(5)))
 
list(chain(range(4),'abc'))
 
list(chain(('a','b','c'),'nihao',['shijie','zhongguo']))
?
1
2
3
4
5
def chain(*iterables):
 # chain('abc', 'def') --> a b c d e f
 for it in iterables:
 for element in it:
  yield element

2.compress()

第一个参数为可迭代类型,第二个参数为0和1的集合,两者长度可以不等,

这个暂时不知道可以用在哪里、

?
1
list(compress(['a','b','c','d','e'],[0,1,1,1,0,1]))
?
1
2
3
def compress(data, selectors):
 # compress('abcdef', [1,0,1,0,1,1]) --> a c e f
 return (d for d, s in izip(data, selectors) if s)

3.dropwhile()

接受两个参数,第一个参数为一个判断类似于if语句的函数,丢弃满足的项,直到第一个不满足的项出现时停止丢弃,就是

?
1
2
3
4
5
#伪代码大概是这个样子的
if condition:
 drop element
 while not condition:
 stop drop
?
1
list(dropwhile(lambda x:x>5,range(10,0,-1)))
?
1
2
3
4
5
6
7
8
9
def dropwhile(predicate, iterable):
 # dropwhile(lambda x: x<5, [1,4,6,4,1]) --> 6 4 1
 iterable = iter(iterable)
 for x in iterable:
 if not predicate(x):
  yield x
  break
 for x in iterable:
 yield x

4.groupby

对给定可迭代集合(有重复元素)进行分组,返回的是一个元组,元组的第一个为分组的元素,第二个为分组的元素集合,还是看代码吧:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
for x,y in groupby(['a','a','b','b','b','b','c','d','e','e']):
 print x
 print list(y)
 print ''
 
out:
 a
 ['a', 'a']
 
 b
 ['b', 'b', 'b', 'b']
 
 c
 ['c']
 
 d
 ['d']
 
 e
 ['e', 'e']
?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
class groupby(object):
 # [k for k, g in groupby('aaaabbbccdaabbb')] --> a b c d a b
 # [list(g) for k, g in groupby('aaaabbbccd')] --> aaaa bbb cc d
 def __init__(self, iterable, key=none):
 if key is none:
  key = lambda x: x
 self.keyfunc = key
 self.it = iter(iterable)
 self.tgtkey = self.currkey = self.currvalue = object()
 def __iter__(self):
 return self
 def next(self):
 while self.currkey == self.tgtkey:
  self.currvalue = next(self.it) # exit on stopiteration
  self.currkey = self.keyfunc(self.currvalue)
 self.tgtkey = self.currkey
 return (self.currkey, self._grouper(self.tgtkey))
 def _grouper(self, tgtkey):
 while self.currkey == tgtkey:
  yield self.currvalue
  self.currvalue = next(self.it) # exit on stopiteration
  self.currkey = self.keyfunc(self.currvalue)

5.ifilter()

这个有点像是filter函数了,不过有点不同,filter返回的是一个完成后的列表,而ifilter则是一个生成器,使用的yield

?
1
2
#这样写只是为了更清楚看到输出,其实这么写就跟filter用法一样了,体现不到ifilter的优越之处了
list(ifilter(lambda x:x%2,range(10)))

6.ifilterfalse()

这个跟ifilter用法很像,只是两个是相反数的关系。

?
1
list(ifilterfalse(lambda x:x%2,range(10)))

7.islice()

接受三个参数,可迭代参数,起始切片点,结束切片点,最少给定两个参数,当只有两个参数为默认第二个参数为结束切片点:

?
1
2
3
4
5
in: list(islice(range(10),2,none))
out: [2, 3, 4, 5, 6, 7, 8, 9]
 
in: list(islice(range(10),2))
out: [0, 1]

8.imap()

接受参数个数跟目标函数有关:

?
1
2
3
4
5
6
7
8
#接受两个参数时
list(imap(abs,range(-5,5)))
 
#接受三个参数时
list(imap(pow,range(-5,5),range(10)))
 
#接受四个参数时
list(imap(lambda x,y,z:x+y+z,range(10),range(10),range(10)))

9.starmap()

这个是imap的变异,即只接受两个参数,目标函数会作用在第二个参数集合中、

?
1
2
inlist(starmap(pow,[(1,2),(2,3)]))
out: [1, 8]

10.tee()

接受两个参数,第一个参数为可迭代类型,第二个为int,如果第二个不指定则默认为2,即重复两次,有点像是生成器repeat的生成器类型,

这个就有意思了,是双重生成器输出:

?
1
2
for x in list(tee('abcde',3)):
 print list(x)

11.takewhile()

这个有点跟dropwhile()很是想象,一个是丢弃,一个是拿取:

伪代码为:

?
1
2
3
4
if condition:
 take this element
 while not condition:
 stop take

eg:

?
1
2
inlist(takewhile(lambda x:x<10,(1,9,10,11,8)))
out: [1, 9]

12.izip()

这个跟imap一样,只不过imap是针对map的生成器类型,而izip是针对zip的:

?
1
list(izip('ab','cd'))

13.izip_longest

针对izip只截取最短的,这个是截取最长的,以none来填充空位:

list(izip_longest('a','abcd'))

三、组合迭代器

对python中的高效迭代器函数详解

1.product()

这个有点像是多次使用for循环,两者可以替代。

?
1
2
3
4
5
6
7
list(product(range(10),range(10)))
 
#本质上是这种的生成器模式
l = []
for x in range(10):
 for y in range(10):
 l.append((x,y))

2.permutations()

接受两个参数,第二个参数不设置时输出的没看出来是什么鬼,

第二个参数用来控制生成的元组的元素个数,而输出的元组中最后一个元素是打乱次序的,暂时也不知道可以用在哪

?
1
list(permutations(range(10),2))

3.combinations()

用来排列组合,抽样不放回,第二个参数为参与排列组合的个数

?
1
list(combinations('abc',2))
?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
def combinations(iterable, r):
 # combinations('abcd', 2) --> ab ac ad bc bd cd
 # combinations(range(4), 3) --> 012 013 023 123
 pool = tuple(iterable)
 n = len(pool)
 if r > n:
 return
 indices = range(r)
 yield tuple(pool[i] for i in indices)
 while true:
 for i in reversed(range(r)):
  if indices[i] != i + n - r:
  break
 else:
  return
 indices[i] += 1
 for j in range(i+1, r):
  indices[j] = indices[j-1] + 1
 yield tuple(pool[i] for i in indices)
?
1
2
3
4
5
6
def combinations(iterable, r):
 pool = tuple(iterable)
 n = len(pool)
 for indices in permutations(range(n), r):
 if sorted(indices) == list(indices):
  yield tuple(pool[i] for i in indices)

4.combinations_with_replacement()

与上一个的用法不同的是抽样是放回的

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
def combinations_with_replacement(iterable, r):
 # combinations_with_replacement('abc', 2) --> aa ab ac bb bc cc
 pool = tuple(iterable)
 n = len(pool)
 if not n and r:
 return
 indices = [0] * r
 yield tuple(pool[i] for i in indices)
 while true:
 for i in reversed(range(r)):
  if indices[i] != n - 1:
  break
 else:
  return
 indices[i:] = [indices[i] + 1] * (r - i)
 yield tuple(pool[i] for i in indices)
?
1
2
3
4
5
6
def combinations_with_replacement(iterable, r):
 pool = tuple(iterable)
 n = len(pool)
 for indices in product(range(n), repeat=r):
 if sorted(indices) == list(indices):
  yield tuple(pool[i] for i in indices)

以上这篇对python中的高效迭代器函数详解就是小编分享给大家的全部内容了,希望能给大家一个参考,也希望大家多多支持服务器之家。

原文链接:https://blog.csdn.net/xiaodongxiexie/article/details/54645123

延伸 · 阅读

精彩推荐