20210830 python 在代理ip下通过搜索关键词查询新闻，并通过邮件发送_print(str(i + 1) + '.' + title[i] + '(' + source[i-CSDN博客

# @Time:2021-8-24 10:26
import random
from email.mime.multipart import MIMEMultipart
import requests
import re
import smtplib
from email.mime.text import MIMEText
from email.utils import formataddr
from email.header import Header
import datetime
import time


def baidu(company):
    proxy_list = [
        'http://117.177.250.151:8081',
        'http://111.85.219.250:3129',
        'http://122.70.183.138:8118',
        'http://117.88.21.43:3000',
    ]
    proxy_ip = random.choice(proxy_list)  # 随机获取代理ip
    proxies = {'http': proxy_ip}

    print(proxies)
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36'}
    url = 'https://www.baidu.com/s?tn=news&rtt=1&bsst=1&cl=2&wd=' + company  # 把链接中rtt参数换成4即是按时间排序，默认为1按焦点排序，
    ##问题：可以看看第二页搜索页，网页链接的代码规律；
    ##区别在于第一页什么都没加，第二页最后加了"pn=10",第三页加了"pn=20",因为每一页有10个新闻
    res = requests.get(url, headers=headers, proxies=proxies).text  # 加上headers用来告诉网站这是通过一个浏览器进行的访问

    p_href = '<h3 class="news-title_1YtI1"><a href="(.*?)"'
    href = re.findall(p_href, res, re.S)  # re.S自动考虑源代码换行，爬取新闻链接，一次性会爬取出10个网页链接，每个都存在这个叫"href"的链表里
    p_title = '<h3 class="news-title_1YtI1">.*?>(.*?)</a>'
    title = re.findall(p_title, res, re.S)  ##爬取出新闻标题
    p_date = '<span class="c-color-gray2 c-font-normal">(.*?)</span>'
    date = re.findall(p_date, res)  ##爬取新闻发布日期
    p_source = '<span class="c-color-gray c-font-normal c-gap-right">(.*?)</span>'
    source = re.findall(p_source, res)  ##爬取新闻来源
    #

    news = []
    # for i in range(len(title)):  # range(len(title)),这里因为知道len(title) = 10，所以也可以写成for i in range(10)
    for i in range(10):

        try :
            ...
            if '小时前' in date[i]:
                title[i] = title[i].strip()  # strip()函数用来取消字符串两端的换行或者空格，不过这里好像不太需要了
                title[i] = re.sub('<.*?>', '', title[i])  # 核心，用re.sub()函数来替换不重要的内容
                time.sleep(5)  # 不要爬太快，爬太快会被百度反爬
                print(str(i + 1) + '.' + title[i], source[i], date[i])
                print(href[i])
                news.extend('【' + company + '】' + str(i + 1) + '.' + title[i][0:] + ' ' + source[i] + ' ' + date[i] + '<br>')
                news.extend(href[i][0:] + '<br>')
        except:
            ...
    return news


def send_news():
    date = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')  # 获取当前日期时间
    # 输入Email地址和口令:
    mail_user_name = 'miya'
    mail_user = 'm*****o@163.com'
    mail_pass = 'CE*******G'
    # 输入收件人地址:
    # receivers = ["33******@qq.com","m*******.com"]
    receivers = ["33*******@qq.com"]
    # 输入SMTP服务器地址:
    smtp_server = 'smtp.163.com'

    # companys = ['光器件', '光芯片']
    # companys = ['5G市场','波分复用器','800G光模块','平面波导','8寸晶圆','WDM']
    companys = ['半导体', '光模块', '5G', '光纤', '5G市场', '运营商', '通信', '晶圆', '光器件', '光芯片', '光缆', '光通信', '光迅科技', '中际旭创', '华工科技', '富通鑫茂', '中环股份', '华为', '新易盛', '中国移动', '中国电信', '中国联通']

    news = '<html><body>'
    news = news + "<h2>今日行业新闻采集汇总:</h2>" + date + "<br>"+ "<br>"
    msg_header = '今日行业新闻自动采集 - miya'
    for i in companys:  # 这个i只是个代号，可以换成其他内容
        news_list = baidu(i)

        #输入要发送的内容:
        # news_list = get_news()
        # print(news_list)
        msg_header = '今日行业新闻自动收集统计 - miya'

        for li in range(0,len(news_list)):
            news = news + news_list[li]

        news = news + "<hr style='background-color:red; height:2px; border:none;'/>"

    # print(news)

    news = news + '</body></html>'

    msg = MIMEMultipart('alternative')
    msg.attach(MIMEText('hello', 'plain', 'utf-8'))
    msg.attach(MIMEText(news, 'html', 'utf-8'))


    # msg = MIMEText(news, 'html', 'utf-8')
    msg['From'] = formataddr([mail_user_name, mail_user])
    server = smtplib.SMTP_SSL(smtp_server)
    msg['Subject'] = Header(msg_header, 'utf-8')
    # SMTP协议默认端口是25
    # server.set_debuglevel(1)
    server.login(mail_user, mail_pass)
    try:
        for item in receivers:
            server.sendmail(mail_user, [item], msg.as_string())
        print("Success: 已成功发送邮件!")
        server.quit()
    except smtplib.SMTPException:
        print("Error: 无法发送邮件")


if __name__ == '__main__':
    send_news()