zoukankan      html  css  js  c++  java
  • 京东商城商品价格获取方法(转帖) 枫

    using System;
    using System.Collections.Generic;
    using System.Text;
    using System.Drawing;
    using System.Net;
    using System.Web;
    using HtmlAgilityPack;
    using System.Text.RegularExpressions;
    using System.IO;

    namespace Pmars
    {
        class JingDong
        {
            //得到京东页面的编码格式
            static Encoding encoding = Encoding.GetEncoding("gb2312");

            //如果知道某件商品的ID,我们如何得到这件商品的价格
            public static double GetPrice(string id)
            {
                double price = 0;

                //第一种方式
                //详细查看以前发过的图片识别的博客,利用识别京东价格图片的方法去获得商品的价格
                price = GetPriceByImage(id);

                //第二种方式
                //虽然京东的价格被做成了图片的模式,但是,在购物车或者在结算的时候的价格不是图片的
                //这样我们就可以想办法得到那里的价格就可以了
                //实验证明,在获得购物车或者结算时的页面需要cookie
                //那么,我们可以去下载另一个页面,之后记录cookie就可以解决问题了
                //这里,定义一个类去继承WebClient,用来记录前后的cookie
                price = GetPriceByCookie(id);

                //第三种方式
                //虽然京东在价格上做了很多的手脚,但是我们可以变相的采用其他的方式来处理
                //比如,我们发现,京东有手机商城,这样,我们在手机商场上做了查看
                //发现,手机商城根本就不用Cookie,而是直接下载页面就可以得到数据
                //好吧,看下面的程序就可以了
                price = GetPriceByMobile(id);
                return price;
            }

            private static double GetPriceByImage(string id)
            {
                //京东的商品的Id是独一无二的,也就是说,每个Id标示了一件商品的内容页
                //得到京东商品的下载页面
                string downUrl = "http://www.360buy.com/product/"+id+".html";
                
                //下载京东的商品内容页面,主要是为了下载图片
                byte[] bytes = new WebClient().DownloadData(downUrl);
                
                //通过编码得到页面的内容string,用HttpUtility.HtmlDecode解码
                string content = HttpUtility.HtmlDecode(encoding.GetString(bytes));

                //利用HtmlAgilityPack来加载分析Html页面内容
                HtmlDocument htmlDoc = new HtmlDocument();
                htmlDoc.LoadHtml(content);

                //得到需要下载的价格图片节点
                var imgNode = htmlDoc.DocumentNode.SelectSingleNode(@"//strong[@class='price']/img");

                //得到需要下载的图片地址,需要用当前的页面地址去拼接一下,否则有可能是部分地址
                string imgUrl = new Uri(new Uri(downUrl),imgNode.Attributes["src"].Value).AbsoluteUri;

                //下载图片,放到Images文件夹下
                //生成图片的地址,@"Images/" + 生成一个Guid以保证他们的名字都是不同的 + 图片的格式
                string imgPath = @"Images/" + Guid.NewGuid().ToString().Replace("-", "") + imgUrl.Substring(imgUrl.LastIndexOf('.'));
                //下载图片
                new WebClient().DownloadFile(imgUrl, imgPath);

                //分析图片,得到商品的价格
                JingdongImage jdImage = new JingdongImage();
                string priceStr = jdImage.GetPicNum(imgPath);

                //定义一个匹配double的正则,从价格字符串中得到价格
                //更简单的方法就是priceStr = priceStr.SubString(1);//去掉第一个¥字符
                Regex doublePattern = new Regex(@"\d+(\.\d+)?", RegexOptions.Compiled);
                string dbPrice = doublePattern.Match(priceStr).Value;

                double price = 0;
                if(double.TryParse(dbPrice,out price))
                    return price;
                return 0;
            }

            private static double GetPriceByCookie(string id)
            {
                //京东的商品的Id是独一无二的,也就是说,每个Id标示了一件商品的内容页
                //得到京东商品的加入购物车的链接,用程序来模仿人来将商品加入到购物车里面去
                //加入购物车的链接
                string goUrl = "http://jd2008.360buy.com/purchase/InitCart.aspx?pid=" + id + "&pcount=1&ptype=1";

                //定义一个CookieWebClient类来达到记录Cookie的效果
                CookieWebClient client = new CookieWebClient();

                //去下载一个页面来记录Cookie
                client.DownloadData(goUrl);

                //得到结算页面的页面的Url
                string shopUrl = "http://jd2008.360buy.com/purchase/shoppingcart_pop.aspx";

                //下载得到结算页面的数据
                byte[] bytes = client.DownloadData(shopUrl);
                string content = HttpUtility.HtmlDecode(encoding.GetString(bytes));

                //利用HtmlAgilityPack来加载分析Html页面内容
                HtmlDocument htmlDoc = new HtmlDocument();
                htmlDoc.LoadHtml(content);

                //得到商品价格的节点
                var priceNode = htmlDoc.DocumentNode.SelectSingleNode(@"//span[@class='price'][@id='cartBottom_price']");
                if (priceNode == null)//节点查找失败
                    return 0;

                //得到商品的价格string,去掉第一个字符¥
                string priceStr = priceNode.InnerText.Substring(1);
                double price = 0;
                if (double.TryParse(priceStr, out price))
                    return price;
                return 0;
            }

            private static double GetPriceByMobile(string id)
            {
                //京东的商品的Id是独一无二的,也就是说,每个Id标示了一件商品的内容页
                //得到京东商品的下载页面
                string downUrl = "http://m.360buy.com/cart/add.action?wareId=" + id;

                //下载京东的商品内容页面,主要是为了下载图片
                byte[] bytes = new WebClient().DownloadData(downUrl);

                //注意手机商城里面的Encoding是UTF8的
                //通过编码得到页面的内容string,用HttpUtility.HtmlDecode解码
                string content = HttpUtility.HtmlDecode(Encoding.UTF8.GetString(bytes));

                //利用HtmlAgilityPack来加载分析Html页面内容
                HtmlDocument htmlDoc = new HtmlDocument();
                htmlDoc.LoadHtml(content);

                //得到价格节点
                var priceNode = htmlDoc.DocumentNode.SelectSingleNode(@"//div[@class='p-price']/strong[@class='flk31'][3]");
                if (priceNode == null)//节点查找失败
                    return 0;

                //得到商品的价格string,去掉第一个字符¥
                string priceStr = priceNode.InnerText;
                double price = 0;
                if (double.TryParse(priceStr, out price))
                    return price;
                return 0;
            }
        }

        class CookieWebClient : WebClient
        {
            private CookieContainer m_container = new CookieContainer();

            protected override WebRequest GetWebRequest(Uri address)
            {
                WebRequest request = base.GetWebRequest(address);
                if (request is HttpWebRequest)
                {
                    (request as HttpWebRequest).CookieContainer = m_container;
                }
                return request;
            }
        }

        class MyImage
        {
            //需要进行分析的图片
            private Bitmap bmpobj;
            //需要继承并且重写的数字比对串,在子类里面需要用到这个去比对数字
            public Dictionary<string, char> numDic = new Dictionary<string, char>();

            //得到图片中某一点的灰度数值
            private int GetGrayNumColor(System.Drawing.Color posClr)
            {
                return (posClr.R * 19595 + posClr.G * 38469 + posClr.B * 7472) >> 16;
            }

            //进行灰度处理
            private void GrayByPixels()
            {
                for (int i = 0; i < bmpobj.Height; i++)
                {
                    for (int j = 0; j < bmpobj.Width; j++)
                    {
                        int tmpValue = GetGrayNumColor(bmpobj.GetPixel(j, i));
                        bmpobj.SetPixel(j, i, Color.FromArgb(tmpValue, tmpValue, tmpValue));
                    }
                }
            }

            //得到一个图片的01代码序列
            private string GetSingleBmpCode(Bitmap singlepic, int dgGrayValue)
            {
                Color piexl;
                string code = "";
                for (int posy = 0; posy < singlepic.Height; posy++)
                    for (int posx = 0; posx < singlepic.Width; posx++)
                    {
                        piexl = singlepic.GetPixel(posx, posy);
                        if (piexl.R < dgGrayValue)    // Color.Black )
                            code = code + "1";
                        else
                            code = code + "0";
                    }
                return code;
            }

            //从一个图片里面得到几个分开的数字小图片
            private Bitmap[] GetPicValidByValue(int dgGrayValue)
            {
                List<Bitmap> PicList = new List<Bitmap>();
                Rectangle cloneRect;
                int posx1 = bmpobj.Width, posy1 = bmpobj.Height, posx2 = 0, posy2 = 0;
                bool cut = false; int last = -1, lastx = 0;
                for (int j = 0; j < bmpobj.Width; j++)      //找有效区
                {
                    cut = false;
                    for (int i = 0; i < bmpobj.Height; i++)
                    {
                        int pixelValue = bmpobj.GetPixel(j, i).R;
                        if (pixelValue < dgGrayValue)     //根据灰度值
                        {
                            if (posx1 > j) posx1 = j;
                            if (posy1 > i) posy1 = i;

                            if (posx2 < j) posx2 = j;
                            if (posy2 < i) posy2 = i;
                            cut = true;
                        }
                    };

                    if (cut)
                        continue;
                    if (last + 1 == j)
                    {
                        last++;
                        continue;
                    }

                    cloneRect = new Rectangle(posx1, posy1, posx2 - posx1 + 1, posy2 - posy1 + 1);
                    lastx = j; last = j; posx1 = bmpobj.Width; posy1 = bmpobj.Height; posx2 = 0; posy2 = 0;
                    PicList.Add(bmpobj.Clone(cloneRect, bmpobj.PixelFormat));//复制小块图
                };

                return PicList.ToArray();
            }

            //得到一个图片的数字串
            private string GetPicNumber()
            {
                GrayByPixels(); //灰度处理
                Bitmap[] pics = GetPicValidByValue(128); //得到有效值
                StringBuilder sb = new StringBuilder();
                char c;
                for (int i = 0; i < pics.Length; ++i)
                {
                    string code = GetSingleBmpCode(pics[i], 128);   //得到代码串
                    if (numDic.TryGetValue(code, out c))
                    {
                        sb.Append(c);
                    }
                }
                return sb.ToString();
            }

            //外部调用,得到一个图片的数字串
            public string GetPicNum(Bitmap pic)
            {
                bmpobj = new Bitmap(pic);    //转换为Format32bppRgb
                return GetPicNumber();
            }

            //外部调用,得到一个图片的数字串(重载)
            public string GetPicNum(string fileName)
            {
                bmpobj = new Bitmap(fileName);
                return GetPicNumber();
            }

            //输出在一幅图里面找到的数字,测试或者找到数字比对串时用
            private void TestNumber()
            {
                GrayByPixels(); //灰度处理
                Bitmap[] pics = GetPicValidByValue(128); //得到有效值
                for (int i = 0; i < pics.Length; ++i)
                {
                    string code = GetSingleBmpCode(pics[i], 128);   //得到代码串
                    Console.WriteLine(i);
                    Console.WriteLine(code);
                }
            }

            //外部调用,输出在一幅图里面找到的数字,测试或者找到数字比对串时用
            public void TestNum(Bitmap pic)
            {
                bmpobj = new Bitmap(pic);    //转换为Format32bppRgb
                TestNumber();
            }

            //外部调用,输出在一幅图里面找到的数字,测试或者找到数字比对串时用(重载)
            public void TestNum(string fileName)
            {
                bmpobj = new Bitmap(fileName);
                TestNumber();
            }
        }

        class JingdongImage : MyImage
        {
            public JingdongImage()
            {
                numDic.Add("111100001111011000000110001100001100000110011000000011110000000011110000000001100000000001100000001111111100000001100000000001100000000001100000000001100000000011110000", '¥');
                numDic.Add("001111100011000110110000011110000011110000011110000011110000011110000011110000011011000110001111100", '0');
                numDic.Add("001100111100001100001100001100001100001100001100001100001100111111", '1');
                numDic.Add("011111100110000110110000011000000011000000110000001100000011000000110000001100000011000000111111111", '2');
                numDic.Add("011111100110000110110000011000000011000000110000111100000000110000000011110000011110000110011111100", '3');
                numDic.Add("00000011000000011100000010110000010011000010001100010000110010000011001111111111000000110000000011000000001100", '4');
                numDic.Add("011111111011000000011000000011000000011111100000000110000000011000000011110000011110000110011111100", '5');
                numDic.Add("000111110001100000011000000110000000110111100111000110110000011110000011110000011011000110001111100", '6');
                numDic.Add("111111111000000011000000110000000110000001100000001100000011000000011000000110000000110000001100000", '7');
                numDic.Add("001111100011000110110000011110000011011000110001111100011000110110000011110000011011000110001111100", '8');
                numDic.Add("001111100011000110110000011110000011110000011011000111001111011000000011000000110000001100011111000", '9');
                numDic.Add("111111", '.');
            }
        }
    }

  • 相关阅读:
    【计算机视觉】深度相机(三)--三种方案对比
    【计算机视觉】深度相机(三)--三种方案对比
    【计算机视觉】深度相机(二)--结构光深度测距
    【计算机视觉】深度相机(二)--结构光深度测距
    【计算机视觉】深度相机(一)--TOF总结
    【计算机视觉】深度相机(一)--TOF总结
    【计算机视觉】人脸表情识别技术
    Http报头Accept与Content-Type的区别
    ajax上传文件,并检查文件类型、检查文件大小
    oracle 生成随机数【待整理】
  • 原文地址:https://www.cnblogs.com/mrray/p/2307331.html
Copyright © 2011-2022 走看看