用微信每天自动给好友发送一个景点介绍

本文介绍了一种使用Python的Selenium和BeautifulSoup库进行网页爬取的技术，具体为从旅游网站抓取景点信息，并利用itchat库实现微信消息自动推送。通过创建数据库映射类和操作数据库的方法，实现了景点信息的存储、查询和更新。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

首先在数据库中创建一张表（Spots），列名分别是spot, url, ident.

创建mapping

#spoturl.py
from sqlalchemy import Column, String, INT, and_
from sqlalchemy.ext.declarative import declarative_base
Base = declarative_base()

#创建实体类，对应DB中的字段
class Spots(Base):

    __tablename__  = 'Spots'
    id = Column(INT(), primary_key=True)
    spot = Column(String())   #景点名称
    url = Column(String())    #景点连接
    ident = Column(INT())    #标识符

    def __init__(self, spot, url, ident):
        self.spot = spot
        self.url = url
        self.ident = ident

    def get_equal_filter(self):
        func = and_(Spots.spot == self.spot,
                    Spots.url == self.url,
                    Spots.ident == self.ident)
        return func

连接数据库

#database.py
import sqlalchemy, types
from sqlalchemy import create_engine
from sqlalchemy.orm import sessionmaker

class DataBase():

    def __init__(self):
        self.engine = create_engine('mssql+pymssql://mstftDB:Qual1234@10.233.156.73/testmstdb', echo=True)  #此处填写自己数据库的配置，这只是实例
        DBSession = sessionmaker(bind=self.engine)
        self.session = DBSession()

    def add_item(self, obj):
        self.session.add(obj)
        self.session.commit()

    def query_all(self, target_class):
        result_list = self.session.query(target_class).all()
        return result_list

    def is_in_database(self, obj):
        if(hasattr(obj, "get_equal_filter")):
            obj_filter = obj.get_equal_filter()
            objs_list = self.session.query(obj.__class__).filter(obj_filter).all()
            if(not objs_list):
                return False
            else:
                return True
        else:
            raise Exception("Cannot find \"get_equal_filter\" in class")

    def update_item(self, obj, update_hash):
        database_obj = self.get_database_obj(obj)
        if (not database_obj):
            database_obj = obj
        for each_key in update_hash.keys():
            code = "database_obj.%s=update_hash[\"%s\"]" % (each_key, each_key)
            exec(code)
        self.session.add(database_obj)
        self.session.commit()
        return database_obj

    def get_database_obj(self, obj):
        if(hasattr(obj, "get_equal_filter")):
            obj_filter = obj.get_equal_filter()
            objs_list = self.session.query(obj.__class__).filter(obj_filter).first()
            if(objs_list):
                return objs_list
            else:
                return None
        else:
            raise Exception("Cannot find \"get_equal_filter\" in class" )
            return None

主程序

#spot.py
import requests
import itchat
import time
from random import choice
from bs4 import BeautifulSoup
from selenium import webdriver
from database import DataBase
from spoturl import Spots


class Spot:

    def __init__(self):
        self.db_obj = DataBase()

    def get_soup(self, url, retry_nums=2):
        # options = Options()
        # options.add_argument('-headless')  # 无头参数
        # driver = Firefox(executable_path='C:/YKP/software/geckodriver.exe',
        # firefox_options=options)  # 配了环境变量第一个参数就可以省了，不然传绝对路径
        time.sleep(5)
        try:    #抓取页面涉及到JS，所以采用selenium结合无头浏览器来解析网页
            driver = webdriver.PhantomJS(executable_path='C:/YKP/software/phantomjs-2.1.1-windows/bin/phantomjs.exe')
            driver.get(url)
            html = driver.page_source
            driver.quit()
            soup = BeautifulSoup(html, "html.parser")
        except requests.HTTPError as e:
            print('Download error'+e.__str__())
            soup = None
            if retry_nums > 0:
                if hasattr(e, 'code') and 500 <= e.code <= 600:
                    return self.get_soup(url, retry_nums-1)
        return soup

    def get_city(self, url):    #先获得所有城市的链接
        soup = self.get_soup(url)
        txt_links = soup.findAll("a", {"class", "link"})
        cityurls = {}
        for city in txt_links[1:]:
            cityname = city.text
            url = city['href']
            cityurls[cityname] = url
        return cityurls

    def get_spot(self, url):     #在每个城市中找出热门景点的链接
        addr_dict = self.get_city(url)
        for (addrs, addrsUrl) in addr_dict.items():
            print(addrs)
            time.sleep(1)
            soup = self.get_soup(addrsUrl)
            try:
                spots_line = soup.find("dl", {"class", "line clrfix"})
                spots_tag = spots_line.findAll('dd', {'class', 'tag show'})
            except:
                print("烂地，一个景点都没有！")
            else:
                for spot_tag in spots_tag:
                    spot_a = spot_tag.find('a')
                    spotname = spot_a.text
                    spoturl = spot_a['href']
                    self.add_db(spotname, spoturl, 0)

    def add_db(self, spotname, spoturl, ident):
        user = Spots(spotname, spoturl, ident)
        if not self.db_obj.is_in_database(user):
            print('spotName:' + spotname + '\n' + 'spotUrl:' + spoturl)
            self.db_obj.add_item(user)

    def query_url(self):
        spot_and_urls = self.db_obj.query_all(Spots)
        spot_and_url = choice(spot_and_urls)
        spot_name = spot_and_url.spot
        spot_url = spot_and_url.url
        spot_iden = spot_and_url.ident
        if spot_iden == 0:    #如果景点没有发送过，就执行，发送完的景点将ident变成1
            self.get_content(spot_name, spot_url)
            user = Spots(spot_name, spot_url, spot_iden)
            spot_dict = {}
            spot_dict['spot'] = spot_name
            spot_dict['url'] = spot_url
            spot_dict['ident'] = 1
            self.db_obj.update_item(user, spot_dict)
        else:
            self.query_url()

    def get_content(self, name, url):    #获取每个景点的介绍
        print(url)
        soup = self.get_soup(url)
        try:
            content_box = soup.find('div', {'class', 'e_db_content_box'}).findAll('p')
        except:
            print('无介绍')
        else:
            content = ''
            for con in content_box:
                content = content+con.text+'\n'
            if content == '':
                print('无'+name+'景区详细介绍'+url)
                self.query_url()
            else:
                print(content)
                self.send_mas(content)

    def send_mas(self, content):    #通过itchat模块实现微信自动登录并发送信息给指定好友
        itchat.auto_login(hotReload=True)
        friends = ['好友备注']    #此处应该写自己好友的备注名
        for friend in friends:
            user = itchat.search_friends(friend)
            username = user[0]["UserName"]
            itchat.send(content, username)

if __name__ == '__main__':
    spot = Spot()
    #先执行以下两行，将所有景点链接存储到DB中，执行完后注释掉即可
    # url = 'http://travel.qunar.com/place/'
    # spot.get_spot(url)

    #可通过schedule自定义发送的时间
    # schedule.every().day.at("08:00").do(spot.query_url)
    # schedule.every().day.at("12:00").do(spot.query_url)
    # schedule.every().day.at("20:00").do(spot.query_url)
    # while True:
    #     schedule.run_pending()
    #     time.sleep(1)

    #也可通过第三方软件定时调用这个主程序
    spot.query_url()