迭代器和可迭代协议

Posted 2022-04-20

可迭代对象包括实际序列和按照需求计算的虚拟序列

#手动迭代：iter和next

#python3.0提供了一个内置函数next，它会自动调用一个对象的next方法，给定一个
#可迭代对象x。调用next(x)等同于x,next()

列表以及其他对象不是自身的迭代器，他们可以调用iter来启动迭代

l = [1, 2, 3]
print(iter(l) is l)

False

#print(l.next())

AttributeError: ‘list‘ object has no attribute ‘next‘

l = [1, 2, 3]
i = iter(l)
print(next(i))

1

print(i.next())

2

l = [1, 2, 3]
for x in l:
print(x**2, end=‘ ‘)

1 4 9

l = [1, 2, 3]
i = iter(l)
while True:
try:
j = next(i)
except StopIteration:
break
else:
print(j ** 2, end=‘ ‘)

1 4 9

python2 中迭代方法是x.next, python3是x.next,但next函数确是都可用的。

其他迭代器

d = ‘a‘:1, ‘b‘:2, ‘c‘:3
for k in d.keys():
print(k, d[k])

a 1

b 2

c 3

#python3中，字典自己有个迭代器
i = iter(d)
print(next(i))
print(next(i))
print(next(i))

a

b

c

因此我们不需要调用keys方法来遍历字典，for将使用迭代协议获取键。

for k in d:
print(k, d[k])

a 1

b 2

c 3

shelves和os.popen的结果也是可迭代的

import os
p = os.popen(‘dir‘)
print(p.next())

驱动器 F 中的卷没有标签。

注意py2中popen对象支持p.next方法，py3中支持p.next()方法。

#迭代协议也是需要将结果包装到一个list中才能看到值。可迭代对象一次返回一

个结果，而不是一个列表

a = range(5)
print(a)

range(0, 5)

i = iter(a)
print(next(i))
print(next(i))

0

1

print(list(range(5)))

[0, 1, 2, 3, 4]

列表解析

l = []
for i in [1, 2, 3]:
l.append(i+1)
print(l)

[2, 3, 4]

相比普通的循环，列表解析也可以达到相同的目的，并且更加简洁，运行速度也比

普通的循环快

print([i+1 for i in [1, 2, 3]])

[2, 3, 4]

文件中的列表解析

with open(‘some.py‘, ‘w‘) as f:
f.write(‘import os\n‘)
f.write(‘a = 1\n‘)
f.write(‘b = 2\n‘)
with open(‘some.py‘) as f:
print(f.read())

import os

a = 1

b = 2

f = open(‘some.py‘)
line = f.readlines()
print(line)

[‘import os\n‘, ‘a = 1\n‘, ‘b = 2\n‘]

如果要去掉行末的换行符的话。

print([l.rstrip() for l in line])

[‘import os‘, ‘a = 1‘, ‘b = 2‘]

因为列表解析像for循环语句是一个迭代环境， 我们甚至不用提前打开文件。

l = [line.rstrip() for line in open(‘some.py‘)]
print(l)

[‘import os‘, ‘a = 1‘, ‘b = 2‘]

除此之外。列表解析的表现力也很强，我们可以在迭代时在一个文件上运行任何的

#字符串操作
print([line.upper() for line in open(‘some.py‘)])

[‘IMPORT OS\n‘, ‘A = 1\n‘, ‘B = 2\n‘]

列表解析也可以嵌套for循环和if

lines = [line.rstrip() for line in open(‘some.py‘) if line[0] == ‘a‘]
print(lines)

[‘a = 1‘]

print([x+y for x in ‘abc‘ for y in ‘lmn‘])

[‘al‘, ‘am‘, ‘an‘, ‘bl‘, ‘bm‘, ‘bn‘, ‘cl‘, ‘cm‘, ‘cn‘]

#列表解析，in成员测试，map内置函数，sorted和zip都使用了迭代协议
#当应用文件时，会自动扫描

upper = [line.upper() for line in open(‘some.py‘)]
print(upper)

[‘IMPORT OS\n‘, ‘A = 1\n‘, ‘B = 2\n‘]

print(list(map(str.upper, open(‘some.py‘))))

[‘IMPORT OS\n‘, ‘A = 1\n‘, ‘B = 2\n‘]

print(‘a = 1\n‘ in open(‘some.py‘))

True

print(sorted(open(‘some.py‘)))

[‘a = 1\n‘, ‘b = 2\n‘, ‘import os\n‘]

print(list(zip(open(‘some.py‘), open(‘some.py‘))))

[(‘import os\n‘, ‘import os\n‘), (‘a = 1\n‘, ‘a = 1\n‘), (‘b = 2\n‘, ‘b = 2\n‘)]

print(list(enumerate(open(‘some.py‘))))

[(0, ‘import os\n‘), (1, ‘a = 1\n‘), (2, ‘b = 2\n‘)]

print(list(filter(bool, open(‘some.py‘))))

[‘import os\n‘, ‘a = 1\n‘, ‘b = 2\n‘]

import functools, operator
print(functools.reduce(operator.add, open(‘some.py‘)))

import os

a = 1

b = 2

print(sum([1, 2, 3, 4, 5]))

15

print(any([‘spam‘, ‘‘, ‘mi‘]))

True

print(all([‘spam‘, ‘‘, ‘mi‘]))

False

print(max([1, 2, 3, 4, 5]))

5

print(min([1, 2, 3, 4, 5]))

1

max 和min 也可以应用于文件

print(max(open(‘some.py‘)))

import os

print(min(open(‘some.py‘)))

a = 1

迭代协议也可直接作用于文件

print(list(open(‘some.py‘)))

[‘import os\n‘, ‘a = 1\n‘, ‘b = 2\n‘]

print(tuple(open(‘some.py‘)))

(‘import os\n‘, ‘a = 1\n‘, ‘b = 2\n‘)

print(‘&&‘.join(open(‘some.py‘)))# (牛逼)
#import os
#&&a = 1
#&&b = 2

a, *b = open(‘some.py‘)
print(a)
print(b)
#import os
#
#[‘a = 1\n‘, ‘b = 2\n‘]

解析和集合

print(set(open(‘some.py‘)))

‘b = 2\n‘, ‘a = 1\n‘, ‘import os\n‘

print(line for line in open(‘some.py‘))

‘a = 1\n‘, ‘b = 2\n‘, ‘import os\n‘

print(ix: line for ix, line in enumerate(open(‘some.py‘)))

0: ‘import os\n‘, 1: ‘a = 1\n‘, 2: ‘b = 2\n‘

print(line for line in open(‘some.py‘) if line[0] == ‘a‘)

‘a = 1\n‘

print(ix: line for ix, line in enumerate(open(‘some.py‘)) if line[0] == ‘a‘)

1: ‘a = 1\n‘

def f(a, b, c):
print(a, b, c, sep = ‘&‘)

f(1, 2, 3)

1-2-3

f(*open(‘some.py‘))

import os

&a = 1

&b = 2

x , y = (1, 2), (3, 4)
print(list(zip(x, y)))

[(1, 3), (2, 4)]

a, b = zip(*zip(x, y))
print(a)
print(b)

(1, 2)

(3, 4)

range和其他迭代器不同，rang它不是自己的迭代器，并且他支持在结果上的多个迭代器

他们会记住自己的位置

r = range(3)
i1 = iter(r)
print(next(i1))

0

print(next(i1))

1

i2 = iter(r)
print(next(i2))

0

print(next(i1))

2

而zip，map,filter不支持相同结果上的活跃迭代器

z = zip((1, 2, 3), (10, 11, 12))
i1 = iter(z)
i2 = iter(z)
print(next(i1))

(1, 10)

print(next(i1))

(2, 11)

print(next(i2))

(3, 12)

m = map(abs,(-1, 0, 1))
i1 = iter(m)
i2 = iter(m)
print(next(i1)) # 1
print(next(i1)) # 0
print(next(i1)) # 1
#print(next(i2)) # StopIteration

r = range(3)
r1, r2 = iter(r), iter(r)
print(next(r1)) # 0
print(next(r1)) # 1
print(next(r1)) # 2
print(next(r2)) # 0

通过iter返回一个新的对象来支持多个迭代器，单个迭代器返回自身。

py3中字典中由于keys不在返回一个列表，所以应该首先用list来展示他

或者使用sorted

d = ‘a‘:1, ‘b‘:2, ‘c‘:3
for k in sorted(d.keys()):
print(k, d[k], end=‘ ‘)

a 1 b 2 c 3

for k in sorted(d):
print(k, d[k], end = ‘ ‘)

a 1 b 2 c 3