AutoInfo/base/base_news_data_collation.py

'''
每日从 mongo 数据库, 做新闻汇总,发送到邮箱
'''
import os
import sys

sys.path.append(os.path.join(os.path.abspath(__file__).split('AutoInfo')[0] + 'AutoInfo'))

from pymongo import MongoClient
from datetime import timedelta
import re
from utils.utils import *

config_json = LoadConfig().load_config()
base_project = LoadConfig().get_base_path()

PROJECT_NAME = config_json.get('PROJECT_NAME')
DB_USER = config_json.get('DB_USER')
DB_PASSWORD = config_json.get('DB_PASSWORD')
DB_IP = config_json.get('DB_IP')
DB_PORT = config_json.get('DB_PORT')
MAIL_HOST = config_json.get('MAIL_HOST')
MAIL_USER = config_json.get('MAIL_USER')
MAIL_PASS = config_json.get('MAIL_PASS')
MAIL_SENDER = config_json.get('MAIL_SENDER')
MAIL_RECEIVERS = config_json.get('MAIL_RECEIVERS')
DB_NAME = config_json.get('DB_NAME')  # 确保配置文件中有这个键
MONGO_LINK = f'mongodb://{DB_USER}:{DB_PASSWORD}@{DB_IP}:{DB_PORT}/'.format(**config_json)
now_day = datetime.now().strftime('%Y-%m-%d')  # 获取今天的日期
filter_days = config_json.get('FILTER_DAYS')
filter_keys = config_json.get('FILTER_KEYS')
filter_switch = True


class NewsDataCollation(object):
    def __init__(self):
        # 第三方 SMTP 服务
        self.mail_host = MAIL_HOST  # 设置服务器
        self.mail_user = MAIL_USER  # 用户名
        self.mail_pass = MAIL_PASS  # 口令

        self.sender = MAIL_SENDER
        self.receivers = [MAIL_RECEIVERS]

        self.processed_data = []

    def load_data(self):
        processed_data = []
        # 读取数据
        print('程序正在读取数据')
        client = MongoClient(MONGO_LINK)
        db = client['NEWS']

        # 根据 self.days 获取日期范围
        start_date = (datetime.now() - timedelta(days=filter_days - 1)).strftime('%Y-%m-%d')
        end_date = datetime.now().strftime('%Y-%m-%d')

        # 构造查询条件，匹配日期范围内的日期
        query = {
            "create_datetime": {
                "$regex": f"^{start_date}|{end_date}",
                "$options": "i"  # 使用不区分大小写的匹配
            }
        }

        # 遍历数据库中的所有集合
        for collection_name in db.list_collection_names():
            print(collection_name)
            collection = db[collection_name]
            cursor = collection.find(query)
            for document in cursor:
                if not document.get('title'):
                    continue

                # 检查 'repush_times' 字段是否存在，如果不存在则默认为 5
                repush_times = document.get('repush_times', 5)

                # 减少 repush_times 的值
                new_repush_times = repush_times - 1

                # 更新数据库中的 repush_times 字段
                collection.update_one(
                    {"_id": document['_id']},  # 假设文档中有 _id 字段作为唯一标识
                    {"$set": {"repush_times": new_repush_times}}
                )

                data = self.process_data(document)
                if data:
                    processed_data.append(data)

        # 关闭MongoDB连接
        client.close()
        return processed_data

    def process_data(self, document):
        # 处理数据
        data = {
            "title": document.get('title') or '',
            "context": document.get('context') or '',
            "source_url": document.get('source_url') or '',
            'link': document.get('link') or '',
            "article_type": document.get('article_type') or '',
            "article_source": document.get('article_source') or '',
            "img_url": document.get('img_url') or '',
            'keyword': document.get('keyword') or '',
            "posted_date": document.get('posted_date') or '',
            "create_time": document.get('create_time') or '',
            "create_datetime": document.get('create_datetime') or '',
            "repush_times": document.get('repush_times', 5) - 1
        }

        data['title'] = self.clean_string(data['title'], 'title')
        data['context'] = self.clean_string(data['context'], 'context')

        return data

    def clean_string(self, input_string, text_type):
        # 清除 title 和 context 中的换行符和制表符
        if not isinstance(input_string, str):
            return ''

        # 清除所有空白字符（包括空格、制表符、换行符等）
        cleaned_string = re.sub(r'\s+', '', input_string)

        if len(cleaned_string) > 100:
            cleaned_string = cleaned_string[:100] + '...'

        if text_type == 'context':
            pass

        return cleaned_string

    def send_email(self, processed_data):
        # 发送邮件
        print('准备发送邮件')
        subject = '新闻汇总sub'
        title = '新闻汇总title'
        text = '********************************************************\n'
        for data in processed_data:
            text += '标题: {}\n'.format(data['title'])
            text += '正文: {}\n'.format(data['context'])
            text += '文章地址: {}\n'.format(data['link'])
            text += '类型: {}\n'.format(data['article_type'])
            text += '板块: {}\n'.format(data['article_source'])
            text += '文章时间: {}\n'.format(data['posted_date'])
            text += '获取时间: {}\n'.format(data['create_datetime'])
            text += '********************************************************\n\n'

        message = MIMEText(text, 'plain', 'utf-8')
        message['From'] = Header(title, 'utf-8')
        message['To'] = Header("auto", 'utf-8')
        message['Subject'] = Header(subject, 'utf-8')

        try:
            smtpObj = smtplib.SMTP_SSL(self.mail_host)
            smtpObj.login(self.mail_user, self.mail_pass)
            smtpObj.sendmail(self.sender, self.receivers, message.as_string())
            print("邮件发送成功")
        except smtplib.SMTPException as e:
            print("Error: 无法发送邮件", e)

    def send_email_with_keyword(self, series, keys, processed_data):
        process_send_data = {}
        keys = keys.split('|')
        have_data_keys = []
        for key in keys:
            # print(f'通过关键字: {key} 过滤')  # 用来调试 key 是否正确
            for data in processed_data:
                if key in data['title'] or key in data['context']:
                    # 如果数据里面无 keyword, 用当前 key 替换一下
                    if not data.get('keyword'):
                        data['keyword'] = key

                    if series not in process_send_data:
                        process_send_data[series] = [data]
                    else:
                        process_send_data[series].append(data)

                    # 储存一下有数据的 key, 输出用
                    have_data_keys.append(key)

        if process_send_data:
            print('{}系列, 以下关键字有数据\n{}'.format(series, list(set(have_data_keys))))
            # 发送邮件
            print('程序正在准备发送邮件的数据')
            for key in process_send_data:
                subject = '新闻汇总sub - {}'.format(series)
                title = '新闻汇总title - {}'.format(series)
                text = '********************************************************\n'
                for data in process_send_data[key]:
                    text += '标题: {}\n'.format(data['title'])
                    text += '正文: {}\n'.format(data['context'])
                    text += '文章地址: {}\n'.format(data['link'])
                    text += '类型: {}\n'.format(data['article_type'])
                    text += '板块: {}\n'.format(data['article_source'])
                    text += '关键词: {}\n'.format(key)
                    text += '文章时间: {}\n'.format(data['posted_date'])
                    text += '获取时间: {}\n'.format(data['create_datetime'])
                    text += '********************************************************\n\n'

                message = MIMEText(text, 'plain', 'utf-8')
                message['From'] = Header(title, 'utf-8')
                message['To'] = Header("auto", 'utf-8')
                message['Subject'] = Header(subject, 'utf-8')

                try:
                    smtpObj = smtplib.SMTP_SSL(self.mail_host)
                    smtpObj.login(self.mail_user, self.mail_pass)
                    smtpObj.sendmail(self.sender, self.receivers, message.as_string())
                    print("关键字: {} 的邮件发送成功".format(series))
                except smtplib.SMTPException as e:
                    print("Error: 无法发送邮件", e)

    def main(self):
        # 加载指定天数的所有数据
        processed_data = self.load_data()

        # 如果无数据, 则退出
        if not processed_data:
            print("没有找到任何数据")
            exit(0)

        # 发送一次所有数据的邮件
        # self.send_email(processed_data)

        # # 这里是通过关键词过滤然后再发送邮件
        if filter_switch and filter_keys:
            for series, keys in filter_keys.items():
                self.send_email_with_keyword(series, keys, processed_data)


if __name__ == '__main__':
    NewsDataCollation().main()