使用urllib进行多线程处理？

Question

我创建了一个程序来计算一个网站中某个单词的出现次数，但是为了对此进行优化，我决定使用多线程来同时计算多个网站中一个单词的出现次数！

所以，我使用两种不同的方法：

• 带有线程模块：

`

import threading
from multiprocessing.pool import ThreadPool as Pool
from bs4 import BeautifulSoup, SoupStrainer
from urllib.request import Request, urlopen
from requests_html import HTMLSession
import urllib.parse
import re
import time
import unidecode

remove_prefix = ["google.", "webcache.", ".youtube.", "blogger.com", "pinterest.", "dailymotion", "linkedin.com", ".pdf", "deezer.com", "facebook.", "twitter.", "paroles", ".aliexpress", "dictionnaire", "video"]

keyword = "capital france"
a1 = "paris"
a1 = a1.lower()

values = {'q': keyword}
data = urllib.parse.urlencode(values)
url = "http://www.google.fr/search?" + data
print(url)

session = HTMLSession()
r = session.get(url)

listlink = list(set(r.html.absolute_links))
for c in remove_prefix:
    listlink = [i for i in [item for item in listlink] if not (c in i)]

scorea1 = 0


def compter(number):
    global scorea1
    req = Request(listlink[number], headers={'User-Agent': 'Mozilla/5.0'})
    resp = BeautifulSoup(urlopen(req).read(), "lxml", parse_only=SoupStrainer('body')).get_text().lower()
    scorea1 += sum(1 for _ in re.finditer(r'\b%s\b' % re.escape(unidecode.unidecode(a1)), unidecode.unidecode(resp)))
    print(scorea1)


def process():
    threads = []
    for i in range(5):
        threads.append(threading.Thread(target=compter, args=(i,)))
        threads[-1].start()
    for t in threads:
        t.join()


process()

print(scorea1)

• 使用ThreadPoolExecutor：

`     导入时间     从current.futures导入ThreadPoolExecutor     从bs4导入BeautifulSoup，SoupStrainer     从urllib.request导入请求，urlopen     从request_html导入HTMLSession     导入urllib.parse     汇入     导入unidecode

start = time.time()

remove_prefix = ["google.", "webcache.", ".youtube.", "blogger.com", "pinterest.", "dailymotion", "linkedin.com", ".pdf", "deezer.com", "facebook.", "twitter.", "paroles", ".aliexpress", "dictionnaire", "video"]

keyword = "capital france"
a1 = "paris"
a1 = a1.lower()

values = {'q': keyword}
data = urllib.parse.urlencode(values)
url = "http://www.google.fr/search?" + data

session = HTMLSession()
r = session.get(url)

listlink = list(set(r.html.absolute_links))
for c in remove_prefix:
    listlink = [i for i in [item for item in listlink] if not (c in i)]

scorea1 = 0


def compte(url):
    global scorea1
    req = Request(url, headers={'User-Agent': 'Mozilla/5.0'})
    resp = BeautifulSoup(urlopen(req).read(), "lxml", parse_only=SoupStrainer('body')).get_text().lower()
    scorea1 += sum(1 for _ in re.finditer(r'\b%s\b' % re.escape(unidecode.unidecode(a1)), unidecode.unidecode(resp)))
    print(scorea1)


with ThreadPoolExecutor(max_workers=5) as executor:
    for i in range(5):
        future = executor.submit(compte, listlink[i])

print(scorea1)
print(time.time() - start)

我的问题：

两个方法都需要大约相同的时间来运行，例如2.5秒，但是我想知道是否可以更快地执行此操作，以及在scorea1 > x时如何中断线程？

（PS：如果有什么需要改进的地方，请告诉我）