使用池追加到数组

Question

我正在尝试从 soccerway.com 中抓取数据并检查该页面是否是要玩的已完成游戏/游戏，每个实例都写入单独的csv文件。我正在浏览 10,000 页，因此使用 Pools 编写了它。但是，我从追加函数中获取空列表，并且无法向 csv 文件写入任何内容。

我尝试直接写入文件而不是附加列表，但这给出了不完整的文件

import requests
from bs4 import BeautifulSoup
import time
import numpy as np
import uuid
import time
from multiprocessing import Pool
import sys, os
fixturesA = []
linksA = []
statsA = []
def parse(url):
try:
#print(url)
delays = [0.25,0.5,0.75,1]
delay = np.random.choice(delays)
#time.sleep(delay)
#r = requests.get(url)
r = requests.get(url, timeout = 10)
soup = BeautifulSoup(r.content, "html.parser")
teams = soup.findAll('h3', attrs = {'class' : 'thick'})
homeTeam = teams[0].text.strip()
awayTeam = teams[2].text.strip()
middle = teams[1].text.strip()
dds = soup.findAll('dd')
date = dds[1].text.strip()
gameWeek = dds[2].text.strip()
if ':' not in middle:
middle = middle.split(" - ")
homeGoals = 0
awayGoals = 0
homeGoals = middle[0]
try:
awayGoals = middle[1]
except Exception as e:
homeGoals = "-1"
awayGoals = "-1"
matchGoals = int(homeGoals) + int(awayGoals)
if(matchGoals >= 0):
if(int(homeGoals) > 0 and int(awayGoals) > 0):
btts = "y"
else:
btts = "n"
halfTimeScore = dds[4].text.strip().split(" - ")
firstHalfHomeGoals = halfTimeScore[0]
firstHalfAwayConc = halfTimeScore[0]
firstHalfAwayGoals = halfTimeScore[1]
firstHalfHomeConc = halfTimeScore[1]
firstHalfTotalGoals = int(firstHalfHomeGoals) + int(firstHalfAwayGoals)
secondHalfHomeGoals = int(homeGoals) - int(firstHalfHomeGoals)
secondHalfAwayConc = int(homeGoals) - int(firstHalfHomeGoals)
secondHalfAwayGoals = int(awayGoals) - int(firstHalfAwayGoals)
secondHalfHomeConc = int(awayGoals) - int(firstHalfAwayGoals)
secondHalfTotalGoals = matchGoals - firstHalfTotalGoals
homeTeamContainers = soup.findAll('div', attrs = {'class' : 'container left'})
homeTeamStarting = homeTeamContainers[2]
homeTeamBench = homeTeamContainers[3]
homeTeamYellows = len(homeTeamStarting.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/700/img/events/YC.png' })) + len(homeTeamBench.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/699/img/events/YC.png' }))
homeTeamReds = len(homeTeamStarting.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/700/img/events/RC.png' })) + len(homeTeamBench.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/699/img/events/RC.png' }))
homeTeamCards = homeTeamYellows + homeTeamReds
awayTeamContainers = soup.findAll('div', attrs = {'class' : 'container right'})
awayTeamStarting = awayTeamContainers[2]
awayTeamBench = awayTeamContainers[3]
awayTeamYellows = len(awayTeamStarting.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/700/img/events/YC.png' })) + len(awayTeamBench.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/699/img/events/YC.png' }))
awayTeamReds = len(awayTeamStarting.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/700/img/events/RC.png' })) + len(awayTeamBench.findAll('img', attrs = {'src' : 'https://s1.swimg.net/gsmf/699/img/events/RC.png' }))
awayTeamCards = awayTeamYellows + awayTeamReds
matchCards = homeTeamCards + awayTeamCards
try:
iframe = soup.findAll('iframe')
iframeSrc = iframe[1]['src']
url = 'https://us.soccerway.com/' + iframeSrc
c = requests.get(url,timeout = 10)
soupC = BeautifulSoup(c.content, "html.parser")
cornerContainer = soupC.findAll('td', attrs = {'class' : 'legend left value'})
homeCorners = cornerContainer[0].text.strip()
awayCornersConc = homeCorners
cornerContainer = soupC.findAll('td', attrs = {'class' : 'legend right value'})
awayCorners = cornerContainer[0].text.strip()
homeCornersConc = awayCorners
matchCorners = int(homeCorners) + int(awayCorners)
print("Got Score . " + homeTeam + " vs " + awayTeam+" . " + gameWeek )
statsA.append(homeTeam + "," + awayTeam  + "," + gameWeek + "," + homeGoals + "," + awayGoals + "," + str(matchGoals) + "," + btts + "," + firstHalfHomeGoals + "," + firstHalfHomeConc + "," + firstHalfAwayGoals + "," + firstHalfAwayConc + "," + str(firstHalfTotalGoals) + "," + str(secondHalfHomeGoals) + "," + str(secondHalfHomeConc) + "," + str(secondHalfAwayGoals) + "," + str(secondHalfAwayConc) + "," + str(secondHalfTotalGoals) + "," + str(homeTeamCards) + "," + str(awayTeamCards) + "," + str(matchCards) + "," + homeCorners + "," + awayCorners + "," + homeCornersConc + "," + awayCornersConc + "," + str(matchCorners)+","+dds[0].text.strip() + "n")
return None
except Exception as e:
print("Got Score no corners. " + homeTeam + " vs " + awayTeam+" . " + gameWeek + " NO FRAME")
statsA.append(homeTeam + "," + awayTeam  + "," + gameWeek + "," + homeGoals + "," + awayGoals + "," + str(matchGoals) + "," + btts + "," + firstHalfHomeGoals + "," + firstHalfHomeConc + "," + firstHalfAwayGoals + "," + firstHalfAwayConc + "," + str(firstHalfTotalGoals) + "," + str(secondHalfHomeGoals) + "," + str(secondHalfHomeConc) + "," + str(secondHalfAwayGoals) + "," + str(secondHalfAwayConc) + "," + str(secondHalfTotalGoals) + "," + str(homeTeamCards) + "," + str(awayTeamCards) + "," + str(matchCards) + "," + "" + "," + "" + "," + "" + "," + "" + "," + ""+","+dds[0].text.strip() + "n")
return None
else:
fixturesA.append(homeTeam + "," + awayTeam  + "," + gameWeek + "," + date + "n")
linksA.append(url + "n")
print(homeTeam + " vs " + awayTeam + " at " + middle + " GW:" + gameWeek)
return None
except Exception as e:
exc_type, exc_obj, exc_tb = sys.exc_info()
fname = os.path.split(exc_tb.tb_frame.f_code.co_filename)[1]
print(exc_type, fname, exc_tb.tb_lineno)
linksA.append(url + "n")
print(url)
return None

stats = open('Statsv2.csv','a',encoding='utf-8')
fixtures = open('fixturesv2.csv','w',encoding='utf-8')
with open('links.txt') as f:
content = f.readlines()
content = [x.strip() for x in content]
links = open('links.txt','w')
if __name__ == '__main__':
start_time = time.time()
p = Pool(20)  # Pool tells how many at a time
records = p.map(parse, content)
p.terminate()
p.join()
print("--- %s seconds ---" % (time.time() - start_time))

Answer 1

我假设你运行的是Windows？那么答案是Windows中的多处理会创建副本而不是分叉。因此，您将主进程与列表一起使用，并且您可以使用自己的单独列表集获得工作进程(来自池(。

工作人员很可能正确填写了他们的列表，但主进程中的列表没有获得任何数据，因此保持为空。工人不归还任何东西。因此，当您在主进程中写入文件时，您将获得空文件。

解决此问题的一种简单方法是在主进程和工作线程之间创建管道或队列，以允许线程之间的通信。您也可以使用共享数组，就像它们由多处理类提供的那样，但您需要知道创建过程中的长度。

请参阅文档：多处理

Answer 2

正如@RaJa所指出的，你实际上并没有做父/控制进程可以看到的任何操作。 最简单的方法是从mapPED 函数返回值

例如，parse()可以在末尾返回元组，如下所示：

def parse(url):
# do work
return url, homeTeam, awayTeam, gameWeek, homeGoals, awayGoals # ...

然后，父进程可以接收值并执行有用的操作，例如将它们保存到 CSV 文件：

import csv
with Pool(20) as pool:
records = pool.map(parse, content)
with open('stats.csv', 'w') as fd:
out = csv.writer(fd)
out.writerow([
'url', 'hometeam', 'awayteam',
# and the remaining column names for the header
])
out.writerows(records)