Python雾里看花-list与set十万数据查询性能对比

原理

• 集合为什么能够如此高效，特别是查找、插入和删除操作？
• 这当然和集合内部的数据结构密不可分。
• 不同于其他数据结构，集合的内部结构都是一张哈希表。
• 对集合来说，区别就是哈希表只有单一的元素。

实例1

# -*- coding: utf-8 -*-

import random
import time
num = 100000
listA = [random.randint(1, i) for i in range(1, num)]
setB = set()
while len(setB) < len(listA):
     setB.add(random.randint(1, num))
setA = set(listA)
if __name__ == '__main__':
    """ list """
    t1 = time.time()
    for i in xrange(0, num):
        if i in listA:
            pass
    print time.time() - t1
    """ set 1 """
    t1 = time.time()
    for i in xrange(0, num):
        if i in setA:
            pass
    print time.time() - t1
    """ set 2"""
    t1 = time.time()
    for i in xrange(0, num):
        if i in setB:
            pass
    print time.time() - t1

---

146.096999884
0.018000125885
0.0239999294281

实例2

# -*-encoding:utf8 -*-


# list version
def find_unique_price_by_list(products):
    unique_price_list = []
    for _, _price in products:
        if _price not in unique_price_list:
            unique_price_list.append(_price)
    return len(unique_price_list)


# set version
def find_unique_price_by_set(products):
    unique_price_set = set()
    for _, _price in products:
        unique_price_set.add(_price)
    return len(unique_price_set)


if __name__ == '__main__':
    import time

    ids = [x for x in range(0, 100000)]
    price = [x for x in range(200000, 300000)]
    product_objs = list(zip(ids, price))

    # 计算列表版本的时间
    start_using_list = time.perf_counter()
    find_unique_price_by_list(product_objs)
    end_using_list = time.perf_counter()
    print("time elapse using list: {}".format(end_using_list - start_using_list))

    # 计算集合版本的时间
    start_using_set = time.perf_counter()
    find_unique_price_by_set(product_objs)
    end_using_set = time.perf_counter()
    print("time elapse using set: {}".format(end_using_set - start_using_set))