Python中的黄页刮板不起作用

Question

我正在尝试从黄页中抓取数据。我用过这个刮刀很多次，但最近它停止了。

遇到这个错误

'NoneType' object has no attribute 'group' 0 results found

谁能帮我解决这个问题

我哪里错了？

import requests
import requests_random_user_agent
import urllib.parse
from bs4 import BeautifulSoup
import re
from math import ceil
import csv
import os
import sys
import subprocess
from os import system, name
import time
from tqdm import tqdm



class Scraper:
    def __init__(self,keyword,location):
        self.keyword=keyword
        self.location=location
        self.params = urllib.parse.urlencode({"search_terms": self.keyword, "geo_location_terms": self.location})

    def get_info(self, link):

        try:
            r = requests.get(link)
            html = BeautifulSoup(r.content, "html.parser")
        except:
            return False
        
        try:
            name = html.find('h1').text
        except:
            name = None
            
        try:
            phone = html.find(class_='phone').text
        except:
            phone = None
            
        
        try:

            website = html.find('a',class_='primary-btn website-link')["href"]

            if len(website.split("?")) > 1:
                website = website.split("?")[0]

        except:
            website = None
            

        try:
            email = html.find('a', class_='email-business')["href"].split(":")[1]
        except:
            email=None
        
        try:
            address = html.find('h2',class_='address').text
        except:
            address=None

        return {"name": name, "email": email, "phone": phone, "address": address, "website":website}
    
    def get_num_pages(self):

        try:
            url = f"https://www.yellowpages.com/search?{self.params}"

            response = requests.get(url)

            html = BeautifulSoup(response.content, "html.parser")
            pagination = html.find(class_="pagination")
            
            if not pagination:
                pagination = 1
                links = html.select("a[class='business-name']")

                num_results = 0
            
                for l in links:
                    try:
                        l["data-analytics"]
                        num_results += 1
                    except:
                        continue

                return num_results, pagination
            
            
            num_results = int(re.search('We found(.*?)results',pagination.text).group(1))


            return num_results, int(ceil(int(num_results) / 30))
            
        except Exception as e:
            print(e)
            return False, False

    def get_links(self, page):
        
        try:
            url = f"https://www.yellowpages.com/search?{self.params}&page={page}"

            response = requests.request("GET", url, timeout=10)
            html = BeautifulSoup(response.content, "html.parser")
            links = html.select("a[class='business-name']")

            links_filtered=[]
            
            for l in links:
                try:
                    l["data-analytics"]
                    links_filtered.append(l)
                except:
                    continue

            links_list = []

            for link in links_filtered:
                links_list.append(f"https://www.yellowpages.com{link['href']}")
            

            return links_list
        except Exception as e:
            print(e)
            return []

def open_file(filename):

    try:
        if sys.platform == "win32":
            os.startfile(filename)
        else:
            opener = "open" if sys.platform == "darwin" else "xdg-open"
            subprocess.call([opener, filename])
    except:
        return False

def create_csv(elements):


    row_list = [["Name", "Address", "Phone", "Email", "Website"]]

    for e in elements:
        
        name = e["name"]
        address = e["address"]
        phone = e["phone"]
        email = e["email"]
        website = e["website"]

        row_list.append([name, address, phone, email, website])

    with open('output.csv', 'w', newline='', encoding='utf8') as file:
        writer = csv.writer(file)
        writer.writerows(row_list)

    


def clear():

    # for windows
    if name == 'nt':
        _ = system('cls')

    # for mac and linux(here, os.name is 'posix')
    else:
        _ = system('clear')


def main():

    clear()

    try:

        while True:
            keyword = input("Keyword: ")
            
            if keyword != "":
                break


        while True:
            city = input("City: ")
            if city != "":
                break

        clear()


        scraper = Scraper(keyword, city)

        results, num_pages = scraper.get_num_pages()

        if not results:
            print("0 results found")
            return False

        print(f"{results} results found {keyword} - {city}")

        data = []

        pages = tqdm(range(1, num_pages + 1))

        for page in pages:
            clear()
            try:
                pages.set_description(f"Scraping page {page}/{num_pages}...")


                links = scraper.get_links(page)

                if not (len(links) > 0):
                    continue

                links = tqdm(links)

                for link in links:
                    try:
                        links.set_description(f"Scraping {link}")

                        info = scraper.get_info(link)
                        # print(info)
                        data.append(info)

                        create_csv(data)

                    except:
                        continue

            except:
                continue
        
        

        print("Opening file...")

        open_file("output.csv")

        print("Task finished")

    except:
        return False

if __name__ == "__main__":
    main()

Answer 1

上线失败

num_results = int(re.search('We found(.*?)results',pagination.text).group(1))

通过打开浏览器对搜索结果页面进行非常简单检查，会显示页面上不存在“我们找到 x 个结果”文本。所以re.search 返回None，即使结果很多。

调整您的脚本以在没有 num_pages 的情况下工作，并且仅通过底部的页面链接或通过增加 URL 中的 page= 参数来分页，直到不再列出任何结果/页面。

仅供参考，下次请进行一些最小的调试工作，而不是发布整个脚本。