# _*_ coding : UTF-8 _*_ # @Time : 2024/11/10 21:23 # @UpdateTime : 2024/11/10 21:23 # @Author : haochen zhong # @File : CrawlKejiribao.py # @Software : PyCharm # @Comment : 本程序采集科技日报数据 import asyncio import random from datetime import datetime from bs4 import BeautifulSoup from httpx import AsyncClient from motor.motor_asyncio import AsyncIOMotorClient start_date = datetime.strptime('2011-01', '%Y-%m') """科技日报报2011年1月份开始有数据""" end_date = datetime.today() """截止到今天""" headers = { 'User-Agent': 'User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/107.0.0.0 Safari/537.36 Edg/107.0.1418.42'} # 链接数据库 client = AsyncIOMotorClient('mongodb://localhost:27017') db = client['buweijiguanbao'] collection = db['kejiribao'] async def main(): collection_names = await db.list_collection_names() # 判断数据表是否存在 if "kejiribao" not in collection_names: # 如果不存在，则从2017年9月开始爬取 print("科技日报数据表不存在，开始采集！") await getData(start_date, end_date) else: # 如果存在，则从数据库中获取最后一条记录的日期 last_record = await collection.find_one({}, sort=[('release_time', -1)]) last_date_str = last_record['release_time'] print("数据库截止时间:", last_date_str) await getData(last_date_str, end_date) async def getContent(soup: BeautifulSoup) -> str: """ :param soup: BeautifulSoup对象 :return: 文章内容 """ content = "" for p in soup.select("#ozoom p"): para = p.text.strip() if para: content += para content += '\n' return content async def getData(start_date: datetime, end_date: datetime): """ :param start_date: 开始日期 :param end_date: 结束日期 :return: None """ crawl_num = 0 # 创建一个列表保存月份 months = [] # 从开始日期到结束日期，每个月份都添加到列表中 current_date = start_date while current_date <= end_date: months.append(current_date) # 增加一个月 if current_date.month == 12: current_date = current_date.replace(year=current_date.year + 1, month=1) else: current_date = current_date.replace(month=current_date.month + 1) # 遍历月份列表 for month in months: # 构造URL url = f'https://digitalpaper.stdaily.com/http_www.kjrb.com/kjrb/html/{month.strftime("%Y-%m")}/period.xml' """https://digitalpaper.stdaily.com/http_www.kjrb.com/kjrb/html/2011-10/period.xml""" print(url) async with AsyncClient(headers=headers, timeout=60) as client: response = await client.get(url) response.encoding = response.charset_encoding print(f"一级连接状态：{response.status_code}") if response.status_code == 200: soup = BeautifulSoup(response.text, 'xml') for period in soup.select("period"): period_name = datetime.strptime(period.select_one("period_name").text.strip(), "%Y-%m-%d") front_page = period.select_one("front_page").text.strip() try: url1 = f"https://digitalpaper.stdaily.com/http_www.kjrb.com/kjrb/html/{period_name.strftime('%Y-%m/%d')}/{front_page}" """https://digitalpaper.stdaily.com/http_www.kjrb.com/kjrb/html/2024-10/30/node_2.htm""" print(url1) response2 = await client.get(url1) response2.encoding = response2.charset_encoding print(f"二级连接状态：{response2.status_code}") if response2.status_code == 200: soup2 = BeautifulSoup(response2.text, 'lxml') for item in soup2.select("#pageLink"): banmianming = item.text.split("：")[-1] banmianhao = item.text.split("：")[0] url2 = f"https://digitalpaper.stdaily.com/http_www.kjrb.com/kjrb/html/{period_name.strftime('%Y-%m/%d')}/" + item.get( "href").replace("./", "").strip() print(url2) response3 = await client.get(url2) response3.encoding = response3.charset_encoding print(f"三级连接状态：{response3.status_code}") if response3.status_code == 200: soup3 = BeautifulSoup(response3.text, 'lxml') pages = soup3.select(".ul02_l a") if pages: titleTag = "body > table:nth-child(2) > tbody > tr:nth-child(1) > td:nth-child(2) > table:nth-child(3) > tbody > tr:nth-child(1) > td > table > tbody > tr > td > strong" subTitleTag = "body > table:nth-child(2) > tbody > tr:nth-child(1) > td:nth-child(2) > table:nth-child(3) > tbody > tr:nth-child(1) > td > table > tbody > tr > td > span:nth-child(5)" preTitleTag = "body > table:nth-child(2) > tbody > tr:nth-child(1) > td:nth-child(2) > table:nth-child(3) > tbody > tr:nth-child(1) > td > table > tbody > tr > td > span:nth-child(1)" authorTag = "body > table:nth-child(2) > tbody > tr:nth-child(1) > td:nth-child(2) > table:nth-child(3) > tbody > tr:nth-child(1) > td > table > tbody > tr > td > span:nth-child(7)" else: pages = soup3.select(".title a") titleTag = ".biaoti" subTitleTag = ".futi" preTitleTag = ".yinti" authorTag = ".autor" for item2 in pages: url3 = f"https://digitalpaper.stdaily.com/http_www.kjrb.com/kjrb/html/{period_name.strftime('%Y-%m/%d')}/" + item2.get( "href") """https://digitalpaper.stdaily.com/http_www.kjrb.com/kjrb/html/2024-10/30/content_579753.htm?div=-1""" if await collection.find_one({"detail_url": url3}, {"_id": False}): continue title = item2.text.strip() print(url3) response4 = await client.get(url3) response4.encoding = response4.charset_encoding print(f"四级连接状态：{response4.status_code}") if response4.status_code == 200: soup4 = BeautifulSoup(response4.text, 'lxml') try: title = soup4.select_one(titleTag).text.strip() except: title = title try: subTitle = soup4.select(subTitleTag)[0].text.strip() except: subTitle = "" try: author = soup4.select_one(authorTag).text.strip() except: author = "" try: perTitle = soup4.select(preTitleTag)[-1].text.strip() except: perTitle = "" try: keywordlist = soup4.find("founder-keyword").text.strip() except: keywordlist = "" content = await getContent(soup4) await collection.insert_one({ "title": title, "subtitle": subTitle, "preTitle": perTitle, "author": author, "banmianming": banmianming, "banmianhao": banmianhao, 'keywordlist': keywordlist, 'detail_url': url3, 'release_time': period_name, 'insert_timestamp': datetime.today(), 'content': content }) crawl_num += 1 print( f"科技日报---{period_name.strftime('%Y-%m-%d')}----{banmianming}---{banmianhao}---{title}---采集完成！") await asyncio.sleep(random.randint(5, 15)) print( f"科技日报---{period_name.strftime('%Y-%m-%d')}----{banmianming}---{banmianhao}-----采集完成！") await asyncio.sleep(random.randint(5, 15)) print(f"科技日报---{period_name.strftime('%Y-%m-%d')}-----采集完成！") await asyncio.sleep(random.randint(5, 15)) except Exception as e: print(e) await collection.insert_one( {'banmianhao': 'empty', 'banmianming': 'empty', 'preTitle': 'empty', 'title': 'empty', 'subtitle': 'empty', 'author': 'empty', 'keywordlist': 'empty', 'detail_url': url, 'release_time': period_name, 'insert_timestamp': datetime.today(), 'content': 'empty'} ) print(f"科技日报采集完毕，共采集{crawl_num}条数据！") asyncio.run(main())