从网站抓取网页以捕获分页链接的问题

Question

我正在尝试从主页（已完成）上所有列出的类别 URL 以及网站及其分页链接的更多子类别页面中抓取数据。网址是here

我已经创建了 Python 脚本来提取模块化结构中的数据，因为我需要在一个单独的文件中从一个步骤到另一个步骤的所有 URL 的输出。但是现在我面临着提取所有分页 URL 的问题，之后将从中获取数据。此外，我仅从第一个子类别 URL 获取数据，而不是来自所有列出的子类别 URL 的数据。

例如在我下面的脚本中，数据来自 >>>>>

一般实践（主类别页面）-http://www.medicalexpo.com/cat/general-practice-K.html 和进一步的听诊器（子类别页面）-http://www.medicalexpo.com/medical-manufacturer/stethoscope-2.html

只是来了。我想要此链接上给出的所有列出的子类别链接的数据

如果能从所有列出的子类别页面中获得具有产品 URL 的所需输出，我们将不胜感激。

下面是代码：

import re
import time
import random
import selenium.webdriver.support.ui as ui
from selenium.common.exceptions import TimeoutException, NoSuchElementException 
from selenium import webdriver 
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from lxml import html  
from bs4 import BeautifulSoup
from datetime import datetime
import csv
import os
from fake_useragent import UserAgent

# Function to write data to a file:
def write_to_file(file,mode, data, newline=None, with_tab=None):   #**
    with open(file, mode, encoding='utf-8') as l:
        if with_tab == True:
            data = ''.join(data)
        if newline == True:
            data = data+'\n'
        l.write(data)

# Function for data from Module 1:
def send_link(link1):
    browser = webdriver.Chrome()
    browser.get(link1)
    current_page = browser.current_url 
    print (current_page) 
    soup = BeautifulSoup(browser.page_source,"lxml")
    tree = html.fromstring(str(soup))

# Added try and except in order to skip/pass attributes without any value.
    try:
        main_category_url = browser.find_elements_by_xpath("//li[@class=\"univers-group-item\"]/span/a[1][@href]")
        main_category_url = [i.get_attribute("href") for i in main_category_url[4:]]
        print(len(main_category_url))

    except NoSuchElementException:
        main_category_url = ''

    for index, data in enumerate(main_category_url):
        with open('Module_1_OP.tsv', 'a', encoding='utf-8') as outfile:
            data = (main_category_url[index] + "\n")
            outfile.write(data)

# Data Extraction for Categories under HEADERS:
    try:
        sub_category_url = browser.find_elements_by_xpath("//li[@class=\"category-group-item\"]/a[1][@href]")
        sub_category_url = [i.get_attribute("href") for i in sub_category_url[:]]
        print(len(sub_category_url))
    except NoSuchElementException:
        sub_category_url = ''

    for index, data in enumerate(sub_category_url):
        with open('Module_1_OP.tsv', 'a', encoding='utf-8') as outfile:
            data = (sub_category_url[index] + "\n")
            outfile.write(data)
            
    csvfile = open("Module_1_OP.tsv") 
    csvfilelist = csvfile.readlines()
    send_link2(csvfilelist)

# Function for data from Module 2:
def send_link2(links2): 
    browser = webdriver.Chrome()
    start = 7
    end = 10
    for link2 in (links2[start:end]):    
        print(link2) 

        ua = UserAgent() 
        try:
            ua = UserAgent()
        except FakeUserAgentError:
            pass

        ua.random == 'Chrome'

        proxies = [] 

        t0 = time.time()
        response_delay = time.time() - t0 
        time.sleep(10*response_delay) 
        time.sleep(random.randint(2,5)) 
        browser.get(link2) 
        current_page = browser.current_url 
        print (current_page) 
        soup = BeautifulSoup(browser.page_source,"lxml")
        tree = html.fromstring(str(soup))

        # Added try and except in order to skip/pass attributes without value.
        try:
            product_url = browser.find_elements_by_xpath('//ul[@class=\"category-grouplist\"]/li/a[1][@href]')
            product_url = [i.get_attribute("href") for i in product_url]
            print(len(product_url))
        except NoSuchElementException:
            product_url = ''

        try:
            product_title = browser.find_elements_by_xpath("//ul[@class=\"category-grouplist\"]/li/a[1][@href]") # Use FindelementS for extracting multiple section data
            product_title = [i.text for i in product_title[:]]
            print(product_title)
        except NoSuchElementException:
            product_title = ''
        
        for index, data2 in enumerate(product_title):
            with open('Module_1_2_OP.tsv', 'a', encoding='utf-8') as outfile:
                data2 = (current_page + "\t" + product_url[index] + "\t" + product_title[index] + "\n")
                outfile.write(data2)

        for index, data3 in enumerate(product_title):
            with open('Module_1_2_OP_URL.tsv', 'a', encoding='utf-8') as outfile:
                data3 = (product_url[index] + "\n")
                outfile.write(data3)

        csvfile = open("Module_1_2_OP_URL.tsv")
        csvfilelist = csvfile.readlines()
        send_link3(csvfilelist)

# Function for data from Module 3:
def send_link3(csvfilelist): 
    browser = webdriver.Chrome()
    for link3 in csvfilelist[:3]:
        print(link3) 
        browser.get(link3) 
        time.sleep(random.randint(2,5))
        current_page = browser.current_url 
        print (current_page) 
        soup = BeautifulSoup(browser.page_source,"lxml")
        tree = html.fromstring(str(soup))

        try:
            pagination = browser.find_elements_by_xpath("//div[@class=\"pagination-wrapper\"]/a[@href]")
            pagination = [i.get_attribute("href") for i in pagination]
            print(pagination)

        except NoSuchElementException:
            pagination = ''

        for index, data2 in enumerate(pagination):
            with open('Module_1_2_3_OP.tsv', 'a', encoding='utf-8') as outfile:
                data2 = (current_page + "\n" + pagination[index] + "\n")
                outfile.write(data2)

        dataset = open("Module_1_2_3_OP.tsv") 
        dataset_dup = dataset.readlines() 
        duplicate(dataset_dup)

# Used to remove duplicate records from a List:
def duplicate(dataset):
    dup_items = set()
    uniq_items = []
    for x in dataset:
        if x not in dup_items:
            uniq_items.append(x)
            dup_items.add(x)
            write_to_file('Listing_pagination_links.tsv','w', dup_items, newline=True, with_tab=True)

    csvfile = open("Listing_pagination_links.tsv") 
    csvfilelist = csvfile.readlines()
    send_link4(csvfilelist)

# Function for data from Module 4:
def send_link4(links3):
    browser = webdriver.Chrome()
    for link3 in links3:
      print(link3)
      browser.get(link3) 
      t0 = time.time()
      response_delay = time.time() - t0 
      time.sleep(10*response_delay) 
      time.sleep(random.randint(2,5)) 
      sub_category_page = browser.current_url 
      print (sub_category_page) 
      soup = BeautifulSoup(browser.page_source,"lxml")
      tree = html.fromstring(str(soup))

      # Added try and except in order to skip/pass attributes without value.
      try:
        product_url1 = browser.find_elements_by_xpath('//div[@class=\"inset-caption price-container\"]/a[1][@href]')
        product_url1 = [i.get_attribute("href") for i in product_url1]
        print(len(product_url1))
      except NoSuchElementException:
        product_url1 = ''

      for index, data in enumerate(product_url1):
        with open('Final_Output_' + datestring + '.tsv', 'a', encoding='utf-8') as outfile:
          data = (sub_category_page + "\t" + product_url1[index] + "\n")
          outfile.write(data)

# PROGRAM STARTS EXECUTING FROM HERE...
# Added to attach Real Date and Time field to Output filename
datestring = datetime.strftime(datetime.now(), '%Y-%m-%d-%H-%M-%S') # For filename
#datestring2 = datetime.strftime(datetime.now(), '%H-%M-%S') # For each record

send_link("http://www.medicalexpo.com/")

Answer 1

您实际上根本不需要 Selenium。下面的代码将获取网站上所有内容的类别、子类别和项目链接、名称和描述。

唯一棘手的部分是处理分页的 while 循环。原则是，如果网站上有“下一步”按钮，我们需要加载更多内容。在这种情况下，网站实际上在下一个标签中为我们提供了“下一个”链接，因此很容易迭代，直到没有更多的下一个链接可以检索。

请记住，当您运行此程序时，它可能需要一段时间。还要记住，您可能应该插入睡眠 - 例如在 1 秒 - 在 while 循环中的每个请求之间很好地对待服务器。

这样做会降低你被禁止/类似情况的风险。

import requests
from bs4 import BeautifulSoup
from time import sleep

items_list = [] # list of dictionaries with this content: category, sub_category, item_description, item_name, item_link 

r = requests.get("http://www.medicalexpo.com/")
soup = BeautifulSoup(r.text, "lxml")
cat_items = soup.find_all('li', class_="category-group-item")
cat_items = [[cat_item.get_text().strip(),cat_item.a.get('href')] for cat_item in cat_items]

# cat_items is now a list with elements like this:
# ['General practice','http://www.medicalexpo.com/cat/general-practice-K.html']
# to access the next level, we loop:

for category, category_link in cat_items[:1]:
    print("[*] Extracting data for category: {}".format(category))

    r = requests.get("http://www.medicalexpo.com/cat/general-practice-K.html")
    soup = BeautifulSoup(r.text, "lxml")
    # data of all sub_categories are located in an element with the id 'category-group'
    cat_group = soup.find('div', attrs={'id': 'category-group'})

    # the data lie in 'li'-tags
    li_elements = cat_group.find_all('li')
    sub_links = [[li.a.get('href'), li.get_text().strip()] for li in li_elements]

    # sub_links is now a list og elements like this:
    # ['http://www.medicalexpo.com/medical-manufacturer/stethoscope-2.html', 'Stethoscopes']

    # to access the last level we need to dig further in with a loop
    for sub_category_link, sub_category in sub_links:
        print("  [-] Extracting data for sub_category: {}".format(sub_category))
        local_count = 0
        load_page = True
        item_url = sub_category_link
        while load_page:
            print("     [-] Extracting data for item_url: {}".format(item_url))
            r = requests.get(item_url)
            soup = BeautifulSoup(r.text, "lxml")
            item_links = soup.find_all('div', class_="inset-caption price-container")[2:]
            for item in item_links:
                item_name = item.a.get_text().strip().split('\n')[0]
                item_link = item.a.get('href')
                try:
                    item_description = item.a.get_text().strip().split('\n')[1]
                except:
                    item_description = None
                item_dict = {
                    "category": category,
                    "subcategory": sub_category,
                    "item_name": item_name,
                    "item_link": item_link,
                    "item_description": item_description
                }
                items_list.append(item_dict)
                local_count +=1
            # all itempages has a pagination element
            # if there are more pages to load, it will have a "next"-class
            # if we are on the last page, the will not be a next class and "next_link" will return None
            pagination = soup.find(class_="pagination-wrapper")
            try:
                next_link = pagination.find(class_="next").get('href', None)
            except:
                next_link = None
            # consider inserting a sleep(1) right about here...
            # if the next_link exists it means that there are more pages to load
            # we'll then set the item_url = next_link and the While-loop will continue
            if next_link is not None:
                item_url = next_link
            else:
                load_page = False
        print("      [-] a total of {} item_links extracted for this sub_category".format(local_count))

# this will yield a list of dicts like this one:

# {'category': 'General practice',
#  'item_description': 'Flac duo',
#  'item_link': 'http://www.medicalexpo.com/prod/boso-bosch-sohn/product-67891-821119.html',
#  'item_name': 'single-head stethoscope',
#  'subcategory': 'Stethoscopes'}

# If you need to export to something like excel, uses pandas. Create a DataFrame and simple load it with the list
# pandas can the export the stuff to excel easily...