<?xml version="1.0"encoding="ISO-8859-1"?>
<catalog>
<cd country="USA">
<title>Empire Burlesque</title>
<artist>Bob Dylan</artist>
<price>10.90</price>
</cd>
</catalog>

/catalog/cd/price

/catalog/cd/*

/catalog/*/price

/*/*/price

/catalog/cd[1]

/catalog/cd[last()]

/catalog/cd[price=10.90]

//@country

//cd[@country='UK']

public static string GetWebClient(string url)
{
    string strHTML = "";
    WebClient myWebClient = new WebClient();            
    Stream myStream = myWebClient.OpenRead(url);
    StreamReader sr = new StreamReader(myStream, Encoding.Default);//注意编码
    strHTML = sr.ReadToEnd();
    myStream.Close();
    return strHTML;
}

/// <summary>添加省级-地区-县市 的城市信息，注意 省会城市 标记5</summary>
/// <param name="cityCode">省份代码</param>
public static void ParsePageByArea(String cityCode)
{
    //更加链接格式和省份代码构造URL
    String url = String.Format("http://www.tianqihoubao.com/lishi/{0}.htm", cityCode);
    //下载网页源代码 
    var docText = HtmlHelper.GetWebClient(url);
    //加载源代码，获取文档对象
    var doc = new HtmlDocument(); doc.LoadHtml(docText);
    //更加xpath获取总的对象，如果不为空，就继续选择dl标签
    var res = doc.DocumentNode.SelectSingleNode(@"/html[1]/body[1]/div[1]/div[6]/div[1]/div[1]/div[3]");
    if (res != null)
    {
        var list = res.SelectNodes(@"dl");//选择标签数组
        if (list.Count < 1) return;
        foreach (var item in list)
        {
            var dd = item.SelectSingleNode(@"dd").SelectNodes("a");
            foreach (var node in dd)
            {
                var text = node.InnerText.Trim();
                //拼音代码要从href属性中进行分割提取
                var herf = node.Attributes["href"].Value.Trim().Split('/', '.');
                Console.WriteLine("{0}:{1}", text, herf[herf.Length - 2]);
            }
        }
    }
}

            //自己项目中用到的
            var res = doc.DocumentNode.SelectSingleNode(@"/html[1]/body[1]//div[@id='trends1_cont_2']");

            if (res != null)
            {
                res.Descendants("a").Each(x =>
                {
                    list.Add(new OptionItem()
                    {
                        Text = x.InnerText,
                        Value = url + x.Attributes["href"].Value.TrimStart('.'),
                    });
                });
               
                res.Descendants("span").Where(c => c.Attributes["class"].Value == "date").Each((i, c) =>
                {
                    list[i].Id = int.Parse(c.InnerText.TrimStart('[').TrimEnd(']').Replace("-", ""));
                    list[i].Tag = c.InnerText;
                });

            }

/// <summary>采集单个城市单个月的历史天气数据</summary>
/// <param name="cityCode">城市拼音代码</param>
/// <param name="year">年份</param>
/// <param name="month">月份</param>
public static void ParsePageByCityMonth(String cityCode, Int32 year, Int32 month)
{
    //更加拼音代码，月份信息构造URL
    String url = String.Format("http://www.tianqihoubao.com/lishi/{0}/month/{1}{2:D2}.html", cityCode, year, month);
    //获取该链接的源代码
    var docText = HtmlHelper.GetWebClient(url);
    //加载源代码，获取页面结构对象
    var doc = new HtmlDocument(); doc.LoadHtml(docText);
    //更加Xpath获取表格对象
    var res = doc.DocumentNode.SelectSingleNode(@"/html[1]/body[1]/div[2]/div[6]/div[1]/div[1]/table[1]");
    if (res != null)
    {
        //获取所有行
        var list = res.SelectNodes(@"tr");
        list.RemoveAt(0);//移除第一行，是表头
        // 遍历每一行，获取日期，以及天气状况等信息
        foreach (var item in list)
        {
            var dd = item.SelectNodes(@"td");
            //日期 -  - 气温 - 风力风向
            if (dd.Count != 4) continue;
            //获取当前行日期
            var date1 = dd[0].InnerText.Replace("\r\n", "").Replace(" ", "").Trim();  
            //获取当前行天气状况
            var tq = dd[1].InnerText.Replace("\r\n", "").Replace(" ", "").Trim();
            //获取当前行气温
            var qw = dd[2].InnerText.Replace("\r\n", "").Replace(" ", "").Trim();
            //获取当前行风力风向
            var fx = dd[3].InnerText.Replace("\r\n", "").Replace(" ", "").Trim();
            //输出
            Console.WriteLine("{0}:{1},{2},{3}", date1, tq, qw, fx);
        }
    }
}