在抓取站点(200或403)时与状态代码不一致

Question

我试图把seekingalpha.com新闻部分作为一个个人项目。然而，我似乎无法成功地模拟浏览器，因为一旦我到达第8页左右，我就得到了403 forbidden output code。如果我在私有模式下打开浏览器，我能够手动浏览所有的页面，所以我的IP不会被阻塞。

我正在使用Requests和Beautifulsoup在Python3.8中

我有：

user-agents

• Using

• 添加了一个合法的用户代理，并尝试了随机的请求会话，这些会话应该是自动更新cookies的，我相信(？)
• 在请求

之间添加了一个引用header

• Increased时延。

这是我的代码：

import requests
import time
import random
import webbrowser
from bs4 import BeautifulSoup
import re
import sys
import os



class SeekingAlpha():

    from fake_useragent import UserAgent
    ua = UserAgent()

    BASE_URL = 'https://seekingalpha.com/'
    NEWS_URL = BASE_URL + 'articles?page={}'


    def __init__(self):
             self.session = requests.Session()
             self.session.headers['User-Agent'] = 'Mozilla/5.0 (X11;  Ubuntu; Linux i686; rv:52.0) Gecko/20100101 Firefox/52.0'

             response =self.session.get(self.BASE_URL)
             response.raise_for_status() 
             self.session.headers['Referrer'] = 'https://seekingalpha.com/'
             print(self.session.headers)
             self.master_urls = []

             for i in range(1,100):        
                page = self.session.get(self.NEWS_URL.format(i))
                time.sleep(random.randint(3,5))
                page.raise_for_status()
                soup = BeautifulSoup(page.content, 'html.parser')
                links = soup.find_all('a', href = True)
                links = [link for link in links if link.has_attr("sasource") and link['sasource'] == 'all_articles']
                self.master_urls.extend(links) 


if __name__ == "__main__":

    master_urls = SeekingAlpha()

编辑：

下面是我通过浏览器看到的第8页的内容(删除标题，以免占用post中太多的空间)：

“最新文章

重点：

    All
    Top Ideas
    Editors' Picks
    Small-Cap Insight
    Outstanding Contribution
    Most Popular

文章/新闻记录

Should I Open A Roth IRA Right Now? That Depends
Charles Lewis Sizemore, CFA • Thu, Apr. 30, 11:15 AM
China Continues To Lead World's Major Equity Regions In 2020
James Picerno • MCHI, SPY, VT• Thu, Apr. 30, 11:09 AM
Gold And Gas: 2 Anti-Recession Trades
Atlas Research • QQQ, UNG, SAND• Thu, Apr. 30, 11:05 AM
Excellent Total Return Bond Funds For Momentum-Based Fixed Income Portfolios
MyPlanIQ • TGMNX, BOND, DLTNX• Thu, Apr. 30, 11:04 AM
NXP's Share Price Already Assumes A Lot Of Growth And Improvement
Stephen Simpson, CFA • MCHP, RNECY, TXN• Thu, Apr. 30, 11:01 AM
[This article is one of the editors' picks] Chart Industries Worth Another Look With LNG Mostly Washed Out
Stephen Simpson, CFA • GTLS• Thu, Apr. 30, 10:53 AM
Dana Incorporated 2020 Q1 - Results - Earnings Call Presentation
SA Transcripts • DAN• Thu, Apr. 30, 10:43 AM
Don't Panic! Coronavirus, GDP, And Unemployment
CFA Institute Contributors • SPY, QQQ, DIA• Thu, Apr. 30, 10:42 AM
Predicting Depressions For Dummies, Part II
John Overstreet • SPY, QQQ, DIA• Thu, Apr. 30, 10:37 AM
Cognex Already Trading On Recovery Prospects
Stephen Simpson, CFA • FANUY, CGNX• Thu, Apr. 30, 10:29 AM
Meritor, Inc. 2020 Q2 - Results - Earnings Call Presentation
SA Transcripts • MTOR• Thu, Apr. 30, 10:28 AM

“

Answer 1

你试过增加随机睡眠吗？我想3-5太低了，网站可能会在你第八次请求之后关闭你。要么增加它，要么如果你得到403去睡觉，然后再试一次。

如果您确实需要该数据，请尽快配置一个Tor代理，并使用它一段时间。(为您提供了一个不同的外部IP -为以防万一而放弃您的会话)

有时候，如果你的机器人变得太烦人，网站的所有者会把你赶出去(至少，这是我的经验:-/ )。