task = [‘任務1', ‘任務2', ‘任務3', …]

from multiprocessing.dummy import Pool as ThreadPool
import os
import requests
import time
import numpy as np
# 文件夾位置
filepath = r'C:\Users\Administrator\Desktop\ceshi'
pool = ThreadPool(10)#開啟線程數,即一次性拋出的請求數
time_list = []#用來計算時間
xml_list = []#數據集
pathDir = os.listdir(filepath)
for i, allDir in enumerate(pathDir):
 filename = os.path.join('%s%s' % (filepath + '\\', allDir))
 kk = open(filename, 'r', encoding='utf-8').read()
 data = kk.encode('utf-8')
 for k in range(10):
  xml_list.append(data)
def res(data):
 # 訪問目標服務器地址
 url_host = 'https://mp.csdn.net/mdeditor#'
 start = time.clock()
 s = requests.post(url_host, data=data)
 end = time.clock()
 if s.status_code == 200:
  print(end-start)
  time_list.append(end-start)
 else:
  print('請求失敗')
# 傳入的參數,1為函數, 2為參數
result = pool.map(res, xml_list)
all_arr = np.array(time_list)
aver = np.mean(all_arr)
variance = np.var(all_arr)
mid = np.median(all_arr)
min_num = np.min(all_arr)
max_num = np.max(all_arr)
print('平均值 : '+ str(aver))
print('方差 : ' + str(variance))
print('中值 : ' + str(mid))
print('最小值 : ' + str(min_num))
print('最大值 : ' + str(max_num))

from multiprocessing.dummy import Pool
poop = Pool(4)  # 4代表電腦是多少核的
results = pool.map(爬取函數,網址列表)

from multiprocessing.dummy import Pool as ThreadPool
import requests
import time
kv = {'user-agent':'Mozilla/5.0'}
def getsource(url):
 html = requests.get(url,headers=kv)
urls = []
for i in range(0,41):
 i = i*50
 newpage = 'https://tieba.baidu.com/f?kw=讀書&ie=utf-8&pn=' + str(i)
 urls.append(newpage)
# 單線程爬取
time1 = time.time()
for each in urls:
 print(each)
 getsource(each)
time2 = time.time()
print('單線程耗時: ' + str(time2-time1))
# 多線程爬取
pool = ThreadPool(8)
time3 = time.time()
results = pool.map(getsource, urls)
pool.close()
pool.join()
time4 = time.time()
print('多線程所消耗時間:' + str(time4 - time3))