爬虫实战--人民网

前言

为了巩固所学的知识，作者尝试着开始发布一些学习笔记类的博客，方便日后回顾。当然，如果能帮到一些萌新进行新技术的学习那也是极好的。作者菜菜一枚，文章中如果有记录错误，欢迎读者朋友们批评指正。
（博客的参考源码可以在我主页的资源里找到，如果在学习的过程中有什么疑问欢迎大家在评论区向我提出）

发现宝藏

前些天发现了一个巨牛的人工智能学习网站，通俗易懂，风趣幽默，忍不住分享一下给大家。【宝藏入口】。

http://jhsjk.people.cn/testnew/result

import os
import re
from datetime import datetime
import requests
import json
from bs4 import BeautifulSoup
from pymongo import MongoClient
from tqdm import tqdm

class ArticleCrawler:
    def __init__(self, catalogues_url, card_root_url, output_dir, db_name='ren-ming-wang'):
        self.catalogues_url = catalogues_url
        self.card_root_url = card_root_url
        self.output_dir = output_dir
        self.client = MongoClient('mongodb://localhost:27017/')
        self.db = self.client[db_name]
        self.catalogues = self.db['catalogues']
        self.cards = self.db['cards']
        self.headers = {
            'Referer': 'https://jhsjk.people.cn/result?',
            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) '
                          'Chrome/119.0.0.0 Safari/537.36',
            'Cookie': '替换成你自己的',
        }

    # 发送带参数的get请求并获取页面内容
    def fetch_page(self, url, page):
        params = {
            'keywords': '',
            'isFuzzy': '0',
            'searchArea': '0',
            'year': '0',
            'form': '',
            'type': '0',
            'page': page,
            'origin': '全部',
            'source': '2',
        }
        response = requests.get(url, params=params, headers=self.headers)
        soup = BeautifulSoup(response.text, 'html.parser')
        return soup

    # 解析请求版面
    def parse_catalogues(self, json_catalogues):
        card_list = json_catalogues['list']
        for list in card_list:
            a_tag = 'article/'+list['article_id']
            card_url = self.card_root_url + a_tag
            card_title = list['title']
            updateTime = list['input_date']
            self.parse_cards(card_url, updateTime)
            date = datetime.now()
            catalogues_id = list['article_id']+'01'

            # 检查重复标题
            existing_docs = self.catalogues.find_one({'id': catalogues_id})
            if existing_docs is not None:
                print(f'版面id: {catalogues_id}【已经存在】')
                continue

            card_data = {
                'id': catalogues_id,
                'title': card_title,
                'page': 1,
                'serial': 1,
                # 一个版面一个文章
                'dailyId': '',
                'cardSize': 1,
                'subjectCode': '50',
                'updateTime': updateTime,
                'institutionnCode': '10000',
                'date': date,
                'snapshot': {

                }
            }
            self.catalogues.insert_one(card_data)
            print(f'版面id: {catalogues_id}【插入成功】')

    # 解析请求文章
    def parse_cards(self, url, updateTime):
        response = requests.get(url, headers=self.headers)
        soup = BeautifulSoup(response.text, "html.parser")
        try:
            title = soup.find("div", "d2txt clearfix").find('h1').text
        except:
            try:
                title = soup.find('h1').text
            except:
                print(f'【无法解析该文章标题】{url}')
        html_content = soup.find('div', 'd2txt_con clearfix')
        text = html_content.get_text()
        imgs = [img.get('src') or img.get('data-src') for img in html_content.find_all('img')]
        cleaned_content = self.clean_content(text)
        # 假设我们有一个正则表达式匹配对象match
        match = re.search(r'\d+', url)
        # 获取匹配的字符串
        card_id = match.group()
        date = datetime.now()
        if len(imgs) != 0:
            # 下载图片
            self.download_images(imgs, card_id)

        # 创建文档
        document = {
            'id': card_id,
            'serial': 1,
            'page': 1,
            'url' : url,
            'type': 'ren-ming-wang',
            'catalogueId': card_id + '01',
            'subjectCode': '50',
            'institutionCode': '10000',
            'updateTime': updateTime,
            'flag': 'true',
            'date': date,
            'title': title,
            'illustrations': imgs,
            'html_content': str(html_content),
            'content': cleaned_content
        }
        # 检查重复标题
        existing_docs = self.cards.find_one({'id': card_id})
        if existing_docs is None:
            # 插入文档
            self.cards.insert_one(document)
            print(f"文章id：{card_id}【插入成功】")
        else:
            print(f"文章id：{card_id}【已经存在】")

    # 文章数据清洗
    def clean_content(self, content):
        if content is not None:
            content = re.sub(r'\r', r'\n', content)
            content = re.sub(r'\n{2,}', '', content)
            # content = re.sub(r'\n', '', content)
            content = re.sub(r' {6,}', '', content)
            content = re.sub(r' {3,}\n', '', content)
            content = content.replace('
', '').replace('<\P>', '').replace(' ', ' ')
        return content

    # 下载图片
    def download_images(self, img_urls, card_id):
        # 根据card_id创建一个新的子目录
        images_dir = os.path.join(self.output_dir, card_id)
        if not os.path.exists(images_dir):
            os.makedirs(images_dir)
            downloaded_images = []
            for img_url in img_urls:
                try:
                    response = requests.get(img_url, stream=True)
                    if response.status_code == 200:
                        # 从URL中提取图片文件名
                        image_name = os.path.join(images_dir, img_url.split('/')[-1])
                        # 确保文件名不重复
                        if os.path.exists(image_name):
                            continue
                        with open(image_name, 'wb') as f:
                            f.write(response.content)
                        downloaded_images.append(image_name)
                        print(f"Image downloaded: {img_url}")
                except Exception as e:
                    print(f"Failed to download image {img_url}. Error: {e}")
            return downloaded_images
        # 如果文件夹存在则跳过
        else:
            print(f'文章id为{card_id}的图片文件夹已经存在')

    # 查找共有多少页
    def find_page_all(self, soup):
        # 查找标签
        em_tag = soup.find('em', onclick=True)
        # 从onclick属性中提取页码
        if em_tag and 'onclick' in em_tag.attrs:
            onclick_value = em_tag['onclick']
            page_number = int(onclick_value.split('(')[1].split(')')[0])
            return page_number
        else:
            print('找不到总共有多少页数据')

    # 关闭与MongoDB的连接
    def close_connection(self):
        self.client.close()

    # 执行爬虫，循环获取多页版面及文章并存储
    def run(self):
        soup_catalogue = self.fetch_page(self.catalogues_url, 1)
        page_all = self.find_page_all(soup_catalogue)
        if page_all:
            for index in tqdm(range(1, page_all), desc='Page'):
            # for index in tqdm(range(1, 50), desc='Page'):
                soup_catalogues = self.fetch_page(self.catalogues_url, index).text
                # 解析JSON数据
                soup_catalogues_json = json.loads(soup_catalogues)
                self.parse_catalogues(soup_catalogues_json)
                print(f'======================================Finished page {index}======================================')

        self.close_connection()

if __name__ == "__main__":
    crawler = ArticleCrawler(
        catalogues_url='http://jhsjk.people.cn/testnew/result',
        card_root_url='http://jhsjk.people.cn/',
        output_dir='D:\\ren-ming-wang\\img'
    )
    crawler.run()  # 运行爬虫，搜索所有内容
    crawler.close_connection()  # 关闭数据库连接