python 高级编程与异步IO并发编程(六)深入python的set和dict

6-1 dict的abc继承关系

from collections import Mapping,MutableMapping
#dict属于mapping类型
a={}
print (isinstance(a,MutableMapping))

6-2 dict的常用方法

a=dict()
a={"bobby1":{"company":"imooc"},"bobby1":{"company":"imooc"}}
#clear
#a.clear()
#pass

#copy,返回浅拷贝
new_dict = a.copy()
new_dict["bobby1"]["company"]="imooc3"

#copy,返回深拷贝
import copy
new_dict=copy.deepcopy(a)
new_dict["bobby1"]["company"]="imooc3"

#formkeys
new_list=["bobby1","bobby2"]

new_dict=dict.fromkeys(new_list,{"company":imooc})
value=new_dict.get("bobby",{})

6-3 dict的子类

#不建议继承set和dict

class Mydict(dict):
    def __setitem__(self,key,value):
        super().__setitem__(key,value*2)
        
my_dict=Mydict(one=1)
my_dict["one"] =1
print(my_dict)

from collections import UserDict

class Mydict(UserDict):
    def __setitem__(self,key,value):
        super().__setitem__(key,value*2)
        
my_dict=Mydict(one=1)
#my_dict["one"] =1
print(my_dict)

from collections import defaultdict 
my_dict=defaultdict(dict)
my_value=my_dict["bobby"]
pass

6-4 set和frozenset

#set 集合 fronzenset(不可变集合)无序，不重复

s = set('abcde')
s = set(['a','b','c','d','e'])
s={'a','b'}
print(type(s))

s = fornzenset('abcde') #fronzenset 可以作为dict的key

#向set添加数据
s.add()
another_set=set("def")
s.update(another_set)
re_set=s.defference(another_set)
re_set = s-another_set
# / & - #集合运算
#set性能很高
print (s.issubset(re_set))
if "c" in re_set:
    print("i am in set")

6-5 dict和set的实现原理

from random import randint


def load_list_data(total_nums, target_nums):
    """
    从文件中读取数据，以list的方式返回
    :param total_nums: 读取的数量
    :param target_nums: 需要查询的数据的数量
    """
    all_data = []
    target_data = []
    file_name = "G:/慕课网课程/AdvancePython/fbobject_idnew.txt"
    with open(file_name, encoding="utf8", mode="r") as f_open:
        for count, line in enumerate(f_open):
            if count < total_nums:
                all_data.append(line)
            else:
                break

    for x in range(target_nums):
        random_index = randint(0, total_nums)
        if all_data[random_index] not in target_data:
            target_data.append(all_data[random_index])
            if len(target_data) == target_nums:
                break

    return all_data, target_data

def load_dict_data(total_nums, target_nums):
    """
    从文件中读取数据，以dict的方式返回
    :param total_nums: 读取的数量
    :param target_nums: 需要查询的数据的数量
    """
    all_data = {}
    target_data = []
    file_name = "G:/慕课网课程/AdvancePython/fbobject_idnew.txt"
    with open(file_name, encoding="utf8", mode="r") as f_open:
        for count, line in enumerate(f_open):
            if count < total_nums:
                all_data[line] = 0
            else:
                break
    all_data_list = list(all_data)
    for x in range(target_nums):
        random_index = randint(0, total_nums-1)
        if all_data_list[random_index] not in target_data:
            target_data.append(all_data_list[random_index])
            if len(target_data) == target_nums:
                break

    return all_data, target_data


def find_test(all_data, target_data):
    #测试运行时间
    test_times = 100
    total_times = 0
    import time
    for i in range(test_times):
        find = 0
        start_time = time.time()
        for data in target_data:
            if data in all_data:
                find += 1
        last_time = time.time() - start_time
        total_times += last_time
    return total_times/test_times


if __name__ == "__main__":
    all_data, target_data = load_list_data(10000, 1000)
    # all_data, target_data = load_list_data(100000, 1000)
    # all_data, target_data = load_list_data(1000000, 1000)


    # all_data, target_data = load_dict_data(10000, 1000)
    # all_data, target_data = load_dict_data(100000, 1000)
    # all_data, target_data = load_dict_data(1000000, 1000)
    last_time = find_test(all_data, target_data)
    #dict查找的性能远远大于list
    #在list中随着list数据的增大，查找时间会增大
    #在dict中随着dict的增大，查找时间不会增大
    print(last_time)

#dict 的key或者set的值，都必须是可以hash的
#不可变对象 都是可哈希的，str,fronzenset,tuple,自己实现的类，可以实现__hash__方法
#2.dict的内存花销大，但是查询速度快自己定义的对象或者python内部的对象都是用dict包装的
#3.dict的存储顺序和元素的添加顺序有关
#4.添加数据有可能改变已有数据的顺序

6-6 本章小结

python基于协议编程的，dict常用的用法，dict的子类，set和fronzenset,dict背后的哈希表与特性.