广

ASP.NET

  • IOS开发
  • android开发
  • PHP编程
  • JavaScript
  • ASP.NET
  • ASP编程
  • JSP编程
  • Java编程
  • 易语言
  • Ruby编程
  • Perl编程
  • AJAX
  • 正则表达式
  • C语言
  • 编程开发

    asp.net HttpWebRequest自动识别网页编码

    2018-05-08 11:29:41 次阅读 稿源:互联网
    广告

    代码如下:

    static string GetEncoding(string url)
    {
    HttpWebRequest request = null;
    HttpWebResponse response = null;
    StreamReader reader = null;
    try
    {
    request = (HttpWebRequest)WebRequest.Create(url);
    request.Timeout = 20000;
    request.AllowAutoRedirect = false;

    response = (HttpWebResponse)request.GetResponse();
    if (response.StatusCode == HttpStatusCode.OK && response.ContentLength < 1024 * 1024)
    {
    if (response.ContentEncoding != null && response.ContentEncoding.Equals("gzip", StringComparison.InvariantCultureIgnoreCase))
    reader = new StreamReader(new GZipStream(response.GetResponseStream(), CompressionMode.Decompress));
    else
    reader = new StreamReader(response.GetResponseStream(), Encoding.ASCII);

    string html = reader.ReadToEnd();

    Regex reg_charset = new Regex(@"charset/b/s*=/s*(?<charset>[^""]*)");
    if (reg_charset.IsMatch(html))
    {
    return reg_charset.Match(html).Groups["charset"].Value;
    }
    else if (response.CharacterSet != string.Empty)
    {
    return response.CharacterSet;
    }
    else
    return Encoding.Default.BodyName;
    }
    }
    catch
    {
    }
    finally
    {

    if (response != null)
    {
    response.Close();
    response = null;
    }
    if (reader != null)
    reader.Close();

    if (request != null)
    request = null;

    }

    return Encoding.Default.BodyName;
    }

    /// <summary>
    /// 获取源代码
    /// </summary>
    /// <param name="url"></param>
    /// <returns></returns>
    static string GetHtml(string url, Encoding encoding)
    {
    HttpWebRequest request = null;
    HttpWebResponse response = null;
    StreamReader reader = null;
    try
    {
    request = (HttpWebRequest)WebRequest.Create(url);
    request.Timeout = 20000;
    request.AllowAutoRedirect = false;

    response = (HttpWebResponse)request.GetResponse();
    if (response.StatusCode == HttpStatusCode.OK && response.ContentLength < 1024 * 1024)
    {
    if (response.ContentEncoding != null && response.ContentEncoding.Equals("gzip", StringComparison.InvariantCultureIgnoreCase))
    reader = new StreamReader(new GZipStream(response.GetResponseStream(), CompressionMode.Decompress), encoding);
    else
    reader = new StreamReader(response.GetResponseStream(), encoding);
    string html = reader.ReadToEnd();

    return html;
    }
    }
    catch
    {
    }
    finally
    {

    if (response != null)
    {
    response.Close();
    response = null;
    }
    if (reader != null)
    reader.Close();

    if (request != null)
    request = null;

    }

    return string.Empty;
    }

    一起学吧部分文章转载自互联网,供读者交流和学习,若有涉及作者版权等问题请及时与我们联系,以便更正、删除或按规定办理。感谢所有提供资讯的网站,欢迎各类媒体与一起学吧进行文章共享合作。

    广告
    广告
    广告