.net利用爬虫爬取中国天气网的天气信息

1.首先需要添加引用Newtonsoft.Json.dll，这个可以直接从网上获取

2.获取NuGet包HtmlAgilityPack 这是一个非常好用的解析html的工具

3.下载城市编码的数据，下载地址是https://apip.weatherdt.com/float/static/js/city.js  该js直接打开可能会乱码，没关系，在页面上右键，另存为，将它变为txt文件，这时你会发现里面的东西不是乱码了。文件放在bin/debug下，可根据需求自行修改

4.根据城市名称获取城市编码

public string GetCityCode(string cityName)
        {
            //城市编码
            string code = "";
            //读取城市信息文件
            string filePath = AppDomain.CurrentDomain.BaseDirectory + "city.txt";
            FileStream fs = new FileStream(filePath, FileMode.Open);
            StreamReader sr = new StreamReader(fs);
            var json = sr.ReadToEnd();
            sr.Close();
            fs.Close();

            //将获取到的数据解析成json并直接获取第三级城市编码
            JArray jobject = (JArray)JsonConvert.DeserializeObject(json);
            for (int i = 0; i < jobject.Count; i++)
            {
                var arr = (JArray)jobject[i]["children"];
                for (int j = 0; j < arr.Count; j++)
                {
                    var citys = (JArray)arr[j]["children"];
                    
                    for (int m = 0; m < citys.Count; m++)
                    {
                        if (citys[m]["zh"].ToString() == cityName)
                        {
                            code = citys[m]["id"].ToString();
                            break;
                        }
                    }
                    

                }

            }
            return code;
        }

　　

5.获取关键数据

public string InsertWeatherInfo2()
        {
            string result = "0";
            var city = ConfigurationManager.AppSettings["address"]; //解析地址取得想要的信息
            //解析html
            HtmlWeb webClient = new HtmlWeb();
            string cityCode = GetCityCode(city);
            if (cityCode != "")
            {
                try
                {
                    HtmlAgilityPack.HtmlDocument doc = webClient.Load("http://www.weather.com.cn/weather1d/" + cityCode + ".shtml");
                    var json = doc.DocumentNode.SelectSingleNode("//script[contains(text(), 'observe24h_data')]").InnerHtml.Split(';')[0].Replace("var observe24h_data = ", "");

                    string hour_now = DateTime.Now.Hour.ToString();
                    //降水量
                    var jangshui = "0";
                    //气温
                    var qiwen = "";
                    //湿度
                    var shidu = "";
                    //风速
                    var fengsu = "";
                    //风向
                    var fengxiang = "";
                    //监测时间
                    var time = DateTime.Now.ToString("yyyy-MM-dd HH:mm:ss");
                    //guid
                    var guid = Guid.NewGuid().ToString().ToLower();
                    JObject jobject = (JObject)JsonConvert.DeserializeObject(json);
                    //获取所有json中的子集
                    var arr = (JArray)jobject["od"]["od2"];
                   
                    for (int i = 0; i < arr.Count; i++)
                    {
                        //取得当前日期的数据
                        if ((string)arr[i]["od21"] == hour_now)
                        {
                            jangshui = arr[i]["od26"].ToString();
                            qiwen = arr[i]["od22"].ToString();
                            shidu = arr[i]["od27"].ToString();
                            fengsu = arr[i]["od25"].ToString();
                            fengxiang = arr[i]["od24"].ToString();
                            break;
                        }
                    }
                    //向数据库执行插入操作
                  
                }
                catch (Exception ex)
                {
                    result = ex.ToString();
                }
            }
            return result;

        }