用正则表达式自动下载网页中的图片

首先获得网页的Html代码，然后用正则表达式分析其中图片的下载地址，最后自动逐个下载。




Code


using System.Net;






/**///// <summary>


/// Download Image from website


/// better to put into threads


/// </summary>


private void DownloadImage()






{


WebClient c = new WebClient();




//Get html code


string content = c.DownloadString(Uri);




Collection<string> address = new Collection<string>();


Collection<string> name = new Collection<string>();




//Analyse html code to get images address(Uri) and (Name) list


ParseHtml(content, Prefix, PicUriPrefix + @"(?<Uri>[^""]*?)"">(?<Name>[^<]*?)</dt>", address, name);




if (address.Count > 0 && name.Count > 0 && address.Count == name.Count)






{


if (Directory.Exists(folder) == false) Directory.CreateDirectory(folder);//Create folder




foreach (string add in address)






{


//Download images one by one


c.DownloadFile(add, Path.Combine(folder, name[address.IndexOf(add)] + ".jpg"));


}


}


}






/**//// <summary>


/// Parse Html using regular expressions


/// </summary>


/// <param name="content">Html content</param>


/// <param name="prefix">Uri prefix</param>


/// <param name="expression">regular expression</param>


/// <param name="address">Image addresses collection</param>


/// <param name="name">Image names collection</param>


private void ParseHtml(string content, string prefix, string expression, Collection<string> address, Collection<string> name)






{


if (String.IsNullOrEmpty(expression) || address == null || name == null) return;




Regex re = new Regex(expression, RegexOptions.IgnoreCase | RegexOptions.Singleline | RegexOptions.RightToLeft);


MatchCollection mc = re.Matches(content);




if (mc == null || mc.Count == 0) return;




foreach (Match m in mc)






{


address.Add(prefix + m.Groups["Uri"].Value);


name.Add(m.Groups["Name"].Value);


}


}

Html部分代码如下：




Html code


alt="" src="/docfile/dyn/12345678LANGCCCCDDDDEEEEEEEERT09">BMW 3-series</dt>/par


alt="" src="/docfile/dyn/12345678LANGCCCCDDDDEEEEEEEEEE01">Toyota</dt>/par


alt="" src="/docfile/dyn/12345678LANGCCCCDDDDEEEEEEEEJY25">Polo cross</dt>/par


alt="" src="/docfile/dyn/12345678LANGCCCCDDDDEEEEEEEEMO02">Golf 4.5</dt>/par

要用正则表达式从中获得图片下载地址（例如：/docfile/dyn/12345678LANGCCCCDDDDEEEEEEEERT09）和图片名称（例如：BMW 3-series）

正则表达式部分如下：




Regular expression


@"/docfile/dyn/(?<Uri>[^""]*?)"">(?<Name>[^<]*?)</dt>"
注意其中group的用法(? *?)，和引号的匹配用法，匹配双引号前面需要再带一个双引号。其中Uri这个group在dyn/后面，直到双引号结束；后面跟随着双引号和>，然后就是Name这个group；它以<号结束，最后跟随</dt>。