python多线程爬虫框架(1)，自学编程找工作-CSDN博客

本文链接：https://blog.csdn.net/2401_84024148/article/details/137229106

proxy = None

#构造请求头

headers = {“User-agent”:self.user_agent}

#下载页面

result = self.download(url,headers,proxy = proxy,num_retries = self.num_retries)

‘’’

file = open(“f:\\bilibili.html”,“wb”)

file.write(result[“html”])

file.close()

‘’’

if self.cache:

#如果有缓存方式,缓存网页

self.cache[url] = result

print(url,“页面下载完成”)

return result[“html”]

def download(self,url,headers,proxy,num_retries,data=None):

‘’’

用于下载一个页面,返回页面和与之对应的状态码

‘’’

#构建请求

request = urllib.request.Request(url,data,headers or {})

request.add_header(“Cookie”,“finger=7360d3c2; UM_distinctid=15c59703db998-0f42b4b61afaa1-5393662-100200-15c59703dbcc1d; pgv_pvi=653650944; fts=1496149148; sid=bgsv74pg; buvid3=56812A21-4322-4C70-BF18-E6D646EA78694004infoc; CNZZDATA2724999=cnzz_eid%3D214248390-1496147515-https%253A%252F%252Fwww.baidu.com%252F%26ntime%3D1496805293”)

request.add_header(“Upgrade-Insecure-Requests”,“1”)

opener = self.opener or urllib.request.build_opener()

if proxy:

#如果有代理IP,使用代理IP

opener = urllib.request.build_opener(urllib.request.ProxyHandler(proxy))

try:

#下载网页

response = opener.open(request)

print(“code是”,response.code)

html = response.read().decode()

code = response.code

except Exception as e:

print(“下载出现错误”,str(e))

html = ‘’

if hasattr(e,“code”):

code =e.code

if num_retries > 0 and 500<code<600:

#如果错误不是未找到网页,则重新下载num_retries次

return self.download(url,headers,proxy,num_retries-1,data)

else:

code = None

print(html)

return {“html”:html,“code”:code}

class Throttle:

‘’’

按照延时,请求,代理IP等下载网页,处理网页中的link的类

‘’’

def __init__(self, delay):

self.delay = delay

self.domains = {}

def wait(self, url):

‘’’

每下载一个html之间暂停的时间

‘’’

获得域名

domain = urllib.parse.urlparse(url).netloc

获得上次访问此域名的时间

las_accessed = self.domains.get(domain)

if self.delay > 0 and las_accessed is not None:

计算需要强制暂停的时间 = 要求的间隔时间 - (现在的时间 - 上次访问的时间)

sleep_secs = self.delay - (datetime.datetime.now() - las_accessed).seconds

if sleep_secs > 0:

time.sleep(sleep_secs)

存储此次访问域名的时间

self.domains[domain] = datetime.datetime.now()

然后是实现爬虫功能的类

import time

import threading

import re

import urllib.parse

import datetime

from bs4 import BeautifulSoup

from Downloader import Downloader

from MongoCache import MongoCache

SLEEP_TIME = 1

def get_links(html):

‘’’

获得一个页面上的所有链接

‘’’

bs = BeautifulSoup(html, “lxml”)

link_labels = bs.find_all(“a”)

for link in link_labels:

return [link_label.get(‘href’, “default”) for link_label in link_labels]

def same_domain(url1, url2):

‘’’

判断域名书否相同

‘’’

return urllib.parse.urlparse(url1).netloc == urllib.parse.urlparse(url2).netloc

def normalize(seed_url, link):

‘’’

用于将绝对路径转换为相对路径

‘’’

link, no_need = urllib.parse.urldefrag(link)

return urllib.parse.urljoin(seed_url, link)

def threader_crawler(seed_url,resource_regiex=None,link_regiex = “.*”,delay=5,cache=None,download_source_callback=None,user_agent=“wswp”,proxies=None, num_retries=1, max_threads=10, timeout=60,max_url=500):

downloaded = []

crawl_queue = [seed_url]

seen = set([seed_url])

D = Downloader(cache = cache,delay = delay,user_agent=user_agent,proxies=proxies,num_retries=num_retries,timeout=timeout)

print(user_agent)

def process_queue():

while True:

links = []

try:

url = crawl_queue.pop()

except IndexError:

break

else:

html = D(url)

downloaded.append(url)

if download_source_callback:

if resource_regiex and re.match(resource_regiex,url):

download_source_callback(url,html)

links.extend([link for link in get_links(html) if re.match(link_regiex,link)])

for link in links:

link = normalize(seed_url, link)

if link not in seen:

seen.add(link)

if same_domain(seed_url,link):

crawl_queue.append(link)

print(“已经发现的总网页数目为”,len(seen))

print(“已经下载过的网页数目为”,len(downloaded))

print(“还没有遍历过的网页数目为”,len(crawl_queue))

threads=[]

while threads or crawl_queue:

if len(downloaded) == max_url:

return

for thread in threads:

if not thread.is_alive():

threads.remove(thread)

while len(threads) < max_threads and crawl_queue:

print(“线程数量为”, len(threads))

thread = threading.Thread(target=process_queue)

thread.setDaemon(True)

thread.start()

print(“线程数量为”, len(threads))

threads.append(thread)

def main():

starttime = datetime.datetime.now()

threader_crawler(“http://www.xicidaili.com/”,max_threads=1,max_url=10,user_agent=“Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36”)

endtime = datetime.datetime.now()

print(“花费时间”,(endtime-starttime).total_seconds())

if __name__ == “__main__”:

main()

经过测试,多线程爬虫速度要远远高于单个线程爬取,简单测试结果如下

开启30个线程爬取一百个网站用时31秒,平均一个用时0.31秒

开启10个线程爬取一百个网页用时69秒,平均一个用时0.69秒

开启1 个线程爬取一百个网站用时774秒,平均一个用时7.74秒

顺便实现了一个测试用的资源下载类,用于将电影天堂的所有资源页的电影保存到数据库

from lxml import etree

from pymongo import MongoClient

import urllib.request