算法1.1.10 二分查找算法--插入65万记录文件效率与遍历方法比较

在一个65W行数据中插入记录，传统遍历和二分法查询效率差异惊人

DATAS_txt=r'large100W.txt'
# 1.准备文件按照指定行读取
import linecache
import random
# text=linecache.getline(DATAS_txt,1)
import time
 

def write_wenjian111(random_min,random_max,times):
    ff = open(DATAS_txt, "r+")
    words = ff.readlines()
    ff.close()
    for iii in range(len(words)):
        words[iii] = int(words[iii])
    for i in range(times):
        flag=1
        while flag:


            # print(words)

            n=0
            if(len(words)==0):
                flag=0
            a_word=random.randint(random_min, random_max)
            for i in range(len(words)):
                if words[i]==a_word:
                    flag=1
                    break
                elif words[i]>a_word:
                    n=i
                    flag=0
                    break




        words.insert(n,a_word)
    # print(len(words))
    with open(DATAS_txt,"w") as fff:
        for i in words:
            fff.write(str(i)+'\n')

def txt_go111(random_min,random_max,times):
    write_wenjian111(random_min, random_max,times)



def write_wenjian(random_min,random_max,times):
    ff = open(DATAS_txt, "r+")
    words = ff.readlines()
    ff.close()
    # print(words)
    for iii in range(len(words)):
        # print(words[iii])
        # print(iii)
        words[iii] = int(words[iii])

    for i in range(times):

        flag=1
        while flag:



            a_word=random.randint(random_min, random_max)
            firs=0
            lst=len(words)-1
            if firs>lst:
                flag=0

            while firs<=lst:

                mid=int((firs+lst)/2)
                if a_word>words[mid]:
                    firs=mid+1
                    flag = 0
                elif a_word<words[mid]:
                    lst=mid-1
                    flag = 0
                else:
                    flag = 1
                    break

        words.insert(firs,a_word)
    # print(len(words))
    with open(DATAS_txt,"w") as fff:
        for i in words:
            fff.write(str(i)+'\n')


#     生成随机数
#     查找加入

def txt_go(random_min,random_max,times):
    write_wenjian(random_min, random_max,times)




time_start = time.time()
txt_go111(-1000000,1000000,1000)
time_end = time.time()
print('常规TIME:', time_end - time_start, "S")

time_start = time.time()
for i in range(5):
    txt_go(-1000000,1000000,30000)
time_end = time.time()
print('二分TIME:', time_end - time_start, "S")

 

常规遍历插入1000条数据：38.2s

二分法插入150000条数据：30.4s