本文内容整理自《Python Cookbook》,仅用作本人学习笔记,若侵犯原著权益请尽快联系本人。
1.将序列分解为单独变量
- 理解多重赋值
- 丢弃元素方法
data = ('Kevin', 50, 100, (2017, 2, 28))
name, shares, _, date = data
# _ 变量理解为忽略该位置元素
2.从任意长度的可迭代对象中分离元素
- 理解*args为取序列
record = ('Kevin', 50, 100, (2017, 2, 28))
name, *_,(*_,year) = record
# *args会贪婪的获取元祖成员
3.保存最后N个元素
- 掌握collections.deque用法
- deque从两端增删元素复杂度为O(1),普通列表从头部删除元素复杂度为O(N)
from collections import deque
q = deque(maxlen=5)
q.append((1,2))
q.append(4)
q.appendleft(3)
print(q)
q.pop()
q.popleft()
print(q)
##print
# deque([3, (1, 2), 4], maxlen=5)
# deque([(1, 2)], maxlen=5)
4.找到最大或最小的N个元素
- 熟悉堆模块heapq
- heapq.nlargest(n, list, key)
- heapq.nsmallest(n, list, key) #key为可自定义函数
- heapq.heapify(list) #堆化,当list数目较小时比上面快
- max(),min()在取单个元素时较快
- 若待取元素数量接近于列表大小,则先排序再切片性能最好(nlargest和nsmallest也会自动判断这个问题)
5.实现优先级队列
- heapq.heappush(list,item)
- heapq.heappop(list)
- 注意默认生成的堆是以优先级小到大的顺序排序的
import heapq
class priorityList(object):
def __init__(self):
self._queue = []
self._index = 0
def push(self,item,priority):
heapq.heappush(self._queue, (-priority, self._index, item))
self._index += 1
def pop(self):
return heapq.heappop(self._queue)[-1]
if __name__ == '__main__':
q = priorityList()
q.push(5,2)
q.push(3,1)
q.push(7,1)
print(q.pop())
print(q._queue)
6.生成一对多字典
- collections.defaultdict(type)会默认初始化字典key对应的值为“空”
from collections import defaultdict
d = defaultdict(list)
for key, value in pairs:
d[key].append(value)
7.让字典保持有序
- collections.OrderedDict()
from collections import OrderedDict
d = OrderedDict()
##然后当字典用
8.与字典有关的计算问题
- zip(list1,list2)生成一个迭代器内容为list1与list2组合成的元组
min(zip(d.values(), d.keys()))
max(zip(d.values(), d.keys()))
9.两个字典的相同点
- dict.keys()返回keys-view对象
- dict.items()返回item-view对象
- 以上的对象可以进行集合操作 交集 |,差集 -,并集 &
a = {'a':1,'b':2}
c = {key:a[key] for key in a.keys() - ('b')}
# 字典解析
10.序列去重并顺序不变
- set()集合
- yield
- lambda
def depute(items, key=None):
s = set()
for item in items:
val = item if key == None else key(item)
if val not in s:
yield item
s.add(val)
c = [{'x':1,'y':2},{'x':1, 'y':3},{'x':2, 'y':3}]
print(list(depute(c, lambda d:(d['x']))))
11.对切片命名
- slice()可用于切片操作
PRICE = slice(2,4)
record = 'AA123456567899'
cost = int(record[PRICE]) * 2
12.找出序列中出现次数最多的元素
- collections.Counter()
words = ['i','am','i']
word_counts = Counter(words)
print(word_counts.most_common(3))
word_counts['i'] += 1
morewords = ['i','was']
word_counts.update(morewords)
13.通过公共键堆字典列表排序
- operator.itemgetter
from operator import itemgetter
sorted(d, key=itemgetter('lname'))
14.对不原生支持比较的对象排序
- sort的key参数
- lambda
- operator.attrgetter()
15.根据字段将字典列表分组
- itertools.groupby() and sort()
- collections.defaultdict()
from operator import itemgetter
from itertools import groupby
rows.sort(key=itemgetter('date'))
for date, items in groupby(rows, key=itemgetter('date')):
for i in items:
print(i)
16.筛选序列的元素
- 列表推导式[ n if n>10 else 0 for n in xrange(20)]
- 生成器表达式(n if n>10 else 0 for n in xrange(20))
- filter(function, list)
- itertools.compress(可迭代对象, 布尔序列)
17.从字典中提取子集
- 字典推导式{key:value for key,value in prices.items() if value > 200}
18.将名称映射到序列的元素中
- collections.namedtuple()命名元组
from collections import namedtuple
Subscribe = namedtuple('Subscribe',['addr', 'joined'])
sub = Subscribe('[email protected]','2016/1/1')
sub = sub._replace(joined=2)
print(sub.addr)
print(sub.joined)
19.同时对数据做转换和换算
- 在函数参数中使用生成器表达式(函数指sum(),min()等可传入迭代元素函数)
- 注意max(生成器表达式) 比 max( [生成器表达式] ) 省内存
d= [{'name':'kevin', 'price':1},{'name':"openex", 'price':2}]
print(max(s['price'] * 2 for s in d))
###输出: 4
print(max(d, key=lambda s:s['price'] * -1) )
###输出: {'name': 'kevin', 'price': 1}
20.将多个字典映射为一个字典
- collections.ChainMap(*maps)
- map.update(map)
- ChainMap,在修改和查询时默认操作第一个字典