/// <summary> /// 设置代理 /// </summary> /// <param name="item">参数对象</param> private void SetProxy(HttpItem item) { if (!string.IsNullOrWhiteSpace(item.ProxyIp)) { //设置代理服务器 if (item.ProxyIp.Contains(":")) { string[] plist = item.ProxyIp.Split(':'); WebProxy myProxy = new WebProxy(plist[0].Trim(), Convert.ToInt32(plist[1].Trim())); //建议连接 myProxy.Credentials = new NetworkCredential(item.ProxyUserName, item.ProxyPwd); //给当前请求对象 request.Proxy = myProxy; } else { WebProxy myProxy = new WebProxy(item.ProxyIp, false); //建议连接 myProxy.Credentials = new NetworkCredential(item.ProxyUserName, item.ProxyPwd); //给当前请求对象 request.Proxy = myProxy; } request.Credentials = CredentialCache.DefaultCredentials; } else if (item.WebProxy != null) { request.Proxy = item.WebProxy; } }
/// <summary> /// 为请求准备参数 /// </summary> ///<param name="item">参数列表</param> private void SetRequest(HttpItem item) { // 验证证书 SetCer(item); //设置Header参数 if (item.Header != null && item.Header.Count > 0) foreach (string key in item.Header.AllKeys) { request.Headers.Add(key, item.Header[key]); } // 设置代理 SetProxy(item); if (item.ProtocolVersion != null) request.ProtocolVersion = item.ProtocolVersion; request.ServicePoint.Expect100Continue = item.Expect100Continue; //请求方式Get或者Post request.Method = item.Method; request.Timeout = item.Timeout; request.KeepAlive = item.KeepAlive; request.ReadWriteTimeout = item.ReadWriteTimeout; if (!string.IsNullOrWhiteSpace(item.Host)) { request.Host = item.Host; } //Accept request.Accept = item.Accept; //ContentType返回类型 request.ContentType = item.ContentType; //UserAgent客户端的访问类型,包括浏览器版本和操作系统信息 request.UserAgent = item.UserAgent; // 编码 encoding = item.Encoding; //设置Cookie SetCookie(item); //来源地址 request.Referer = item.Referer; //是否执行跳转功能 request.AllowAutoRedirect = item.Allowautoredirect; //设置Post数据 SetPostData(item); //设置最大连接 if (item.Connectionlimit > 0) request.ServicePoint.ConnectionLimit = item.Connectionlimit; }
/// <summary> /// 设置Cookie /// </summary> /// <param name="item">Http参数</param> private void SetCookie(HttpItem item) { if (!string.IsNullOrWhiteSpace(item.Cookie)) //Cookie request.Headers[HttpRequestHeader.Cookie] = item.Cookie; //设置Cookie if (item.ResultCookieType == ResultCookieType.CookieCollection && item.CookieCollection != null && item.CookieCollection.Count > 0) { request.CookieContainer = new CookieContainer(); request.CookieContainer.Add(item.CookieCollection); } }
/// <summary> /// 设置Post数据 /// </summary> /// <param name="item">Http参数</param> private void SetPostData(HttpItem item) { //验证在得到结果时是否有传入数据 if (request.Method.Trim().ToLower().Contains("post")) { if (item.PostEncoding != null) { postencoding = item.PostEncoding; } byte[] buffer = null; //写入Byte类型 if (item.PostDataType == PostDataType.Byte && item.PostdataByte != null && item.PostdataByte.Length > 0) { //验证在得到结果时是否有传入数据 buffer = item.PostdataByte; }//写入文件 else if (item.PostDataType == PostDataType.FilePath && !string.IsNullOrWhiteSpace(item.Postdata)) { StreamReader r = new StreamReader(item.Postdata, postencoding); buffer = postencoding.GetBytes(r.ReadToEnd()); r.Close(); } //写入字符串 else if (!string.IsNullOrWhiteSpace(item.Postdata)) { buffer = postencoding.GetBytes(item.Postdata); } if (buffer != null) { request.ContentLength = buffer.Length; request.GetRequestStream().Write(buffer, 0, buffer.Length); } } }
/// <summary> /// 设置多个证书 /// </summary> /// <param name="item"></param> private void SetCerList(HttpItem item) { if (item.ClentCertificates != null && item.ClentCertificates.Count > 0) { foreach (X509Certificate c in item.ClentCertificates) { request.ClientCertificates.Add(c); } } }
/// <summary> /// 设置证书 /// </summary> /// <param name="item"></param> private void SetCer(HttpItem item) { if (!string.IsNullOrWhiteSpace(item.CerPath)) { //这一句一定要写在创建连接的前面。使用回调的方法进行证书验证。 ServicePointManager.ServerCertificateValidationCallback = new System.Net.Security.RemoteCertificateValidationCallback(CheckValidationResult); //初始化对像,并设置请求的URL地址 request = (HttpWebRequest)WebRequest.Create(item.URL); SetCerList(item); //将证书添加到请求里 request.ClientCertificates.Add(new X509Certificate(item.CerPath)); } else { //初始化对像,并设置请求的URL地址 request = (HttpWebRequest)WebRequest.Create(item.URL); SetCerList(item); } }
/// <summary> /// 根据相传入的数据,得到相应页面数据 /// </summary> /// <param name="item">参数类对象</param> /// <returns>返回HttpResult类型</returns> public HttpResult GetHtml(HttpItem item) { //返回参数 HttpResult result = new HttpResult(); try { //准备参数 SetRequest(item); } catch (Exception ex) { return new HttpResult() { Cookie = string.Empty, Header = null, Html = ex.Message, StatusDescription = "配置参数时出错:" + ex.Message }; } try { #region 得到请求的response using (response = (HttpWebResponse)request.GetResponse()) { result.StatusCode = response.StatusCode; result.StatusDescription = response.StatusDescription; result.Header = response.Headers; if (response.Cookies != null) result.CookieCollection = response.Cookies; if (response.Headers["set-cookie"] != null) result.Cookie = response.Headers["set-cookie"]; byte[] ResponseByte = null; using (MemoryStream _stream = new MemoryStream()) { //GZIIP处理 if (response.ContentEncoding != null && response.ContentEncoding.Equals("gzip", StringComparison.InvariantCultureIgnoreCase)) { //开始读取流并设置编码方式 new GZipStream(response.GetResponseStream(), CompressionMode.Decompress).CopyTo(_stream, 10240); } else { //开始读取流并设置编码方式 response.GetResponseStream().CopyTo(_stream, 10240); } //获取Byte ResponseByte = _stream.ToArray(); } if (ResponseByte != null & ResponseByte.Length > 0) { //是否返回Byte类型数据 if (item.ResultType == ResultType.Byte) result.ResultByte = ResponseByte; //从这里开始我们要无视编码了 if (encoding == null) { Match meta = Regex.Match(Encoding.Default.GetString(ResponseByte), "<meta([^<]*)charset=([^<]*)[\"']", RegexOptions.IgnoreCase); string c = (meta.Groups.Count > 1) ? meta.Groups[2].Value.ToLower().Trim() : string.Empty; if (c.Length > 2) { try { if (c.IndexOf(" ") > 0) c = c.Substring(0, c.IndexOf(" ")); encoding = Encoding.GetEncoding(c.Replace("\"", "").Replace("'", "").Replace(";", "").Replace("iso-8859-1", "gbk").Trim()); } catch { if (string.IsNullOrEmpty(response.CharacterSet)) encoding = Encoding.UTF8; else encoding = Encoding.GetEncoding(response.CharacterSet); } } else { if (string.IsNullOrEmpty(response.CharacterSet)) encoding = Encoding.UTF8; else encoding = Encoding.GetEncoding(response.CharacterSet); } } //得到返回的HTML result.Html = encoding.GetString(ResponseByte); } else { //得到返回的HTML result.Html = "本次请求并未返回任何数据"; } } #endregion } catch (WebException ex) { //这里是在发生异常时返回的错误信息 response = (HttpWebResponse)ex.Response; result.Html = ex.Message; if (response != null) { result.StatusCode = response.StatusCode; result.StatusDescription = response.StatusDescription; } } catch (Exception ex) { result.Html = ex.Message; } if (item.IsToLower) result.Html = result.Html.ToLower(); return result; }
private void CastCont(string link, string FlagCont, string[] DelFlag) { #region 使用HttpHelper取得源码 HttpHelper http = new HttpHelper(); HttpItem item = new HttpItem() { URL = link }; HttpResult result = http.GetHtml(item); string html = result.Html; #endregion #region 使用HtmlAgilityPack解析源码 HtmlAgilityPack.HtmlDocument htmlDocument = new HtmlAgilityPack.HtmlDocument(); htmlDocument.LoadHtml(html); var Nodes = htmlDocument.DocumentNode; #endregion //初始化 rtbCode.Text = html; var reCont = Nodes.CssSelect(FlagCont); foreach (var doc in reCont) { for (int i = 0; i < DelFlag.Length; i++) { foreach (var Del in reCont.CssSelect(DelFlag[i]).ToArray()) Del.Remove(); } htmlEditor1.HTML = doc.InnerHtml; rtbText.Text = doc.InnerText; } }
//截取正文内容部分方法的重构, private void CastCont(string link, string FlagCont) { #region 使用HttpHelper取得源码 HttpHelper http = new HttpHelper(); HttpItem item = new HttpItem() { URL = link }; HttpResult result = http.GetHtml(item); string html = result.Html; #endregion #region 使用HtmlAgilityPack解析源码 HtmlAgilityPack.HtmlDocument htmlDocument = new HtmlAgilityPack.HtmlDocument(); htmlDocument.LoadHtml(html); var Nodes = htmlDocument.DocumentNode; #endregion //初始化 rtbCode.Text = html; var reCont = Nodes.CssSelect(FlagCont); foreach (var doc in reCont) { htmlEditor1.HTML = doc.InnerHtml; rtbText.Text = doc.InnerText; } }