using System; using System.Collections.Generic; using System.IO; using System.Text; using System.Net; using System.Web; namespace Deerchao.Utility { public class HttpClient { #region fields private bool keepContext; private string defaultLanguage = "zh-CN"; private Encoding defaultEncoding = Encoding.UTF8; private string accept = "*/*"; private string userAgent = "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.2; SV1; .NET CLR 1.1.4322; .NET CLR 2.0.50727)"; private HttpVerb verb = HttpVerb.GET; private HttpClientContext context; private readonly List<HttpUploadingFile> files = new List<HttpUploadingFile>(); private readonly Dictionary<string, string> postingData = new Dictionary<string, string>(); private string url; private WebHeaderCollection responseHeaders; private int startPoint; private int endPoint; #endregion #region events public event EventHandler<StatusUpdateEventArgs> StatusUpdate; private void OnStatusUpdate(StatusUpdateEventArgs e) { EventHandler<StatusUpdateEventArgs> temp = StatusUpdate; if (temp != null) temp(this, e); } #endregion #region properties /// <summary> /// 是否自动在不同的请求间保留Cookie, Referer /// </summary> public bool KeepContext { get { return keepContext; } set { keepContext = value; } } /// <summary> /// 期望的回应的语言 /// </summary> public string DefaultLanguage { get { return defaultLanguage; } set { defaultLanguage = value; } } /// <summary> /// GetString()如果不能从HTTP头或Meta标签中获取编码信息,则使用此编码来获取字符串 /// </summary> public Encoding DefaultEncoding { get { return defaultEncoding; } set { defaultEncoding = value; } } /// <summary> /// 指示发出Get请求还是Post请求 /// </summary> public HttpVerb Verb { get { return verb; } set { verb = value; } } /// <summary> /// 要上传的文件.如果不为空则自动转为Post请求 /// </summary> public List<HttpUploadingFile> Files { get { return files; } } /// <summary> /// 要发送的Form表单信息 /// </summary> public Dictionary<string, string> PostingData { get { return postingData; } } /// <summary> /// 获取或设置请求资源的地址 /// </summary> public string Url { get { return url; } set { url = value; } } /// <summary> /// 用于在获取回应后,暂时记录回应的HTTP头 /// </summary> public WebHeaderCollection ResponseHeaders { get { return responseHeaders; } } /// <summary> /// 获取或设置期望的资源类型 /// </summary> public string Accept { get { return accept; } set { accept = value; } } /// <summary> /// 获取或设置请求中的Http头User-Agent的值 /// </summary> public string UserAgent { get { return userAgent; } set { userAgent = value; } } /// <summary> /// 获取或设置Cookie及Referer /// </summary> public HttpClientContext Context { get { return context; } set { context = value; } } /// <summary> /// 获取或设置获取内容的起始点,用于断点续传,多线程下载等 /// </summary> public int StartPoint { get { return startPoint; } set { startPoint = value; } } /// <summary> /// 获取或设置获取内容的结束点,用于断点续传,多下程下载等. /// 如果为0,表示获取资源从StartPoint开始的剩余内容 /// </summary> public int EndPoint { get { return endPoint; } set { endPoint = value; } } #endregion #region constructors /// <summary> /// 构造新的HttpClient实例 /// </summary> public HttpClient() : this(null) { } /// <summary> /// 构造新的HttpClient实例 /// </summary> /// <param name="url">要获取的资源的地址</param> public HttpClient(string url) : this(url, null) { } /// <summary> /// 构造新的HttpClient实例 /// </summary> /// <param name="url">要获取的资源的地址</param> /// <param name="context">Cookie及Referer</param> public HttpClient(string url, HttpClientContext context) : this(url, context, false) { } /// <summary> /// 构造新的HttpClient实例 /// </summary> /// <param name="url">要获取的资源的地址</param> /// <param name="context">Cookie及Referer</param> /// <param name="keepContext">是否自动在不同的请求间保留Cookie, Referer</param> public HttpClient(string url, HttpClientContext context, bool keepContext) { this.url = url; this.context = context; this.keepContext = keepContext; if (this.context == null) this.context = new HttpClientContext(); } #endregion #region AttachFile /// <summary> /// 在请求中添加要上传的文件 /// </summary> /// <param name="fileName">要上传的文件路径</param> /// <param name="fieldName">文件字段的名称(相当于<input type=file name=fieldName>)里的fieldName)</param> public void AttachFile(string fileName, string fieldName) { HttpUploadingFile file = new HttpUploadingFile(fileName, fieldName); files.Add(file); } /// <summary> /// 在请求中添加要上传的文件 /// </summary> /// <param name="data">要上传的文件内容</param> /// <param name="fileName">文件名</param> /// <param name="fieldName">文件字段的名称(相当于<input type=file name=fieldName>)里的fieldName)</param> public void AttachFile(byte[] data, string fileName, string fieldName) { HttpUploadingFile file = new HttpUploadingFile(data, fileName, fieldName); files.Add(file); } #endregion /// <summary> /// 清空PostingData, Files, StartPoint, EndPoint, ResponseHeaders, 并把Verb设置为Get. /// 在发出一个包含上述信息的请求后,必须调用此方法或手工设置相应属性以使下一次请求不会受到影响. /// </summary> public void Reset() { verb = HttpVerb.GET; files.Clear(); postingData.Clear(); responseHeaders = null; startPoint = 0; endPoint = 0; } private HttpWebRequest CreateRequest() { HttpWebRequest req = (HttpWebRequest)WebRequest.Create(url); req.AllowAutoRedirect = false; req.CookieContainer = new CookieContainer(); req.Headers.Add("Accept-Language", defaultLanguage); req.Accept = accept; req.UserAgent = userAgent; req.KeepAlive = false; if (context.Cookies != null) req.CookieContainer.Add(context.Cookies); if (!string.IsNullOrEmpty(context.Referer)) req.Referer = context.Referer; if (verb == HttpVerb.HEAD) { req.Method = "HEAD"; return req; } if (postingData.Count > 0 || files.Count > 0) verb = HttpVerb.POST; if (verb == HttpVerb.POST) { req.Method = "POST"; MemoryStream memoryStream = new MemoryStream(); StreamWriter writer = new StreamWriter(memoryStream); if (files.Count > 0) { string newLine = " "; string boundary = Guid.NewGuid().ToString().Replace("-", ""); req.ContentType = "multipart/form-data; boundary=" + boundary; foreach (string key in postingData.Keys) { writer.Write("--" + boundary + newLine); writer.Write("Content-Disposition: form-data; name="{0}"{1}{1}", key, newLine); writer.Write(postingData[key] + newLine); } foreach (HttpUploadingFile file in files) { writer.Write("--" + boundary + newLine); writer.Write( "Content-Disposition: form-data; name="{0}"; filename="{1}"{2}", file.FieldName, file.FileName, newLine ); writer.Write("Content-Type: application/octet-stream" + newLine + newLine); writer.Flush(); memoryStream.Write(file.Data, 0, file.Data.Length); writer.Write(newLine); writer.Write("--" + boundary + newLine); } } else { req.ContentType = "application/x-www-form-urlencoded"; StringBuilder sb = new StringBuilder(); foreach (string key in postingData.Keys) { sb.AppendFormat("{0}={1}&", HttpUtility.UrlEncode(key), HttpUtility.UrlEncode(postingData[key])); } if (sb.Length > 0) sb.Length--; writer.Write(sb.ToString()); } writer.Flush(); using (Stream stream = req.GetRequestStream()) { memoryStream.WriteTo(stream); } } if (startPoint != 0 && endPoint != 0) req.AddRange(startPoint, endPoint); else if (startPoint != 0 && endPoint == 0) req.AddRange(startPoint); return req; } /// <summary> /// 发出一次新的请求,并返回获得的回应 /// 调用此方法永远不会触发StatusUpdate事件. /// </summary> /// <returns>相应的HttpWebResponse</returns> public HttpWebResponse GetResponse() { HttpWebRequest req = CreateRequest(); HttpWebResponse res = (HttpWebResponse)req.GetResponse(); responseHeaders = res.Headers; if (keepContext) { context.Cookies = res.Cookies; context.Referer = url; } return res; } /// <summary> /// 发出一次新的请求,并返回回应内容的流 /// 调用此方法永远不会触发StatusUpdate事件. /// </summary> /// <returns>包含回应主体内容的流</returns> public Stream GetStream() { return GetResponse().GetResponseStream(); } /// <summary> /// 发出一次新的请求,并以字节数组形式返回回应的内容 /// 调用此方法会触发StatusUpdate事件 /// </summary> /// <returns>包含回应主体内容的字节数组</returns> public byte[] GetBytes() { HttpWebResponse res = GetResponse(); int length = (int)res.ContentLength; MemoryStream memoryStream = new MemoryStream(); byte[] buffer = new byte[0x100]; Stream rs = res.GetResponseStream(); for (int i = rs.Read(buffer, 0, buffer.Length); i > 0; i = rs.Read(buffer, 0, buffer.Length)) { memoryStream.Write(buffer, 0, i); OnStatusUpdate(new StatusUpdateEventArgs((int)memoryStream.Length, length)); } rs.Close(); return memoryStream.ToArray(); } /// <summary> /// 发出一次新的请求,以Http头,或Html Meta标签,或DefaultEncoding指示的编码信息对回应主体解码 /// 调用此方法会触发StatusUpdate事件 /// </summary> /// <returns>解码后的字符串</returns> public string GetString() { byte[] data = GetBytes(); string encodingName = GetEncodingFromHeaders(); if (encodingName == null) encodingName = GetEncodingFromBody(data); Encoding encoding; if (encodingName == null) encoding = defaultEncoding; else { try { encoding = Encoding.GetEncoding(encodingName); } catch (ArgumentException) { encoding = defaultEncoding; } } return encoding.GetString(data); } /// <summary> /// 发出一次新的请求,对回应的主体内容以指定的编码进行解码 /// 调用此方法会触发StatusUpdate事件 /// </summary> /// <param name="encoding">指定的编码</param> /// <returns>解码后的字符串</returns> public string GetString(Encoding encoding) { byte[] data = GetBytes(); return encoding.GetString(data); } private string GetEncodingFromHeaders() { string encoding = null; string contentType = responseHeaders["Content-Type"]; if (contentType != null) { int i = contentType.IndexOf("charset="); if (i != -1) { encoding = contentType.Substring(i + 8); } } return encoding; } private string GetEncodingFromBody(byte[] data) { string encodingName = null; string dataAsAscii = Encoding.ASCII.GetString(data); if (dataAsAscii != null) { int i = dataAsAscii.IndexOf("charset="); if (i != -1) { int j = dataAsAscii.IndexOf(""", i); if (j != -1) { int k = i + 8; encodingName = dataAsAscii.Substring(k, (j - k) + 1); char[] chArray = new char[2] { '>', '"' }; encodingName = encodingName.TrimEnd(chArray); } } } return encodingName; } /// <summary> /// 发出一次新的Head请求,获取资源的长度 /// 此请求会忽略PostingData, Files, StartPoint, EndPoint, Verb /// </summary> /// <returns>返回的资源长度</returns> public int HeadContentLength() { Reset(); HttpVerb lastVerb = verb; verb = HttpVerb.HEAD; using (HttpWebResponse res = GetResponse()) { verb = lastVerb; return (int)res.ContentLength; } } /// <summary> /// 发出一次新的请求,把回应的主体内容保存到文件 /// 调用此方法会触发StatusUpdate事件 /// 如果指定的文件存在,它会被覆盖 /// </summary> /// <param name="fileName">要保存的文件路径</param> public void SaveAsFile(string fileName) { SaveAsFile(fileName, FileExistsAction.Overwrite); } /// <summary> /// 发出一次新的请求,把回应的主体内容保存到文件 /// 调用此方法会触发StatusUpdate事件 /// </summary> /// <param name="fileName">要保存的文件路径</param> /// <param name="existsAction">指定的文件存在时的选项</param> /// <returns>是否向目标文件写入了数据</returns> public bool SaveAsFile(string fileName, FileExistsAction existsAction) { byte[] data = GetBytes(); switch (existsAction) { case FileExistsAction.Overwrite: using (BinaryWriter writer = new BinaryWriter(new FileStream(fileName, FileMode.OpenOrCreate, FileAccess.Write))) writer.Write(data); return true; case FileExistsAction.Append: using (BinaryWriter writer = new BinaryWriter(new FileStream(fileName, FileMode.Append, FileAccess.Write))) writer.Write(data); return true; default: if (!File.Exists(fileName)) { using ( BinaryWriter writer = new BinaryWriter(new FileStream(fileName, FileMode.Create, FileAccess.Write))) writer.Write(data); return true; } else { return false; } } } } public class HttpClientContext { private CookieCollection cookies; private string referer; public CookieCollection Cookies { get { return cookies; } set { cookies = value; } } public string Referer { get { return referer; } set { referer = value; } } } public enum HttpVerb { GET, POST, HEAD, } public enum FileExistsAction { Overwrite, Append, Cancel, } public class HttpUploadingFile { private string fileName; private string fieldName; private byte[] data; public string FileName { get { return fileName; } set { fileName = value; } } public string FieldName { get { return fieldName; } set { fieldName = value; } } public byte[] Data { get { return data; } set { data = value; } } public HttpUploadingFile(string fileName, string fieldName) { this.fileName = fileName; this.fieldName = fieldName; using (FileStream stream = new FileStream(fileName, FileMode.Open)) { byte[] inBytes = new byte[stream.Length]; stream.Read(inBytes, 0, inBytes.Length); data = inBytes; } } public HttpUploadingFile(byte[] data, string fileName, string fieldName) { this.data = data; this.fileName = fileName; this.fieldName = fieldName; } } public class StatusUpdateEventArgs : EventArgs { private readonly int bytesGot; private readonly int bytesTotal; public StatusUpdateEventArgs(int got, int total) { bytesGot = got; bytesTotal = total; } /// <summary> /// 已经下载的字节数 /// </summary> public int BytesGot { get { return bytesGot; } } /// <summary> /// 资源的总字节数 /// </summary> public int BytesTotal { get { return bytesTotal; } } } }
.Net类库里提供了HttpWebRequest等类,方便我们编程与Web服务器进行交互. 但是实际使用中我们经常会遇到以下需求, 基础类里没有直接提供相应的功能 (WebClient类包含这些功能,只是用起来稍微麻烦一点--谢谢网友东吴居士的提醒):
- 对HttpWebResponse获取的HTML进行文字编码转换,使之不会出现乱码;
- 自动在Session间保持Cookie,Referer等相关信息;
- 模拟HTML表单提交;
- 向服务器上传文件;
- 对二进制的资源,直接获取返回的字节数组(byte[]),或者保存为文件
为了解决这些问题,我开发了HttpClient类.下面是使用的方法:
- 获取编码转换后的字符串
HttpClient client=new HttpClient(url);
string html=client.GetString();
GetString()函数内部会查找Http Headers, 以及HTML的Meta标签,试图找出获取的内容的编码信息.如果都找不到,它会使用client.DefaultEncoding, 这个属性默认为utf-8, 也可以手动设置.
- 自动保持Cookie, Referer
HttpClient client=new HttpClient(url1, null, true);
string html1=client.GetString();
client.Url=url2;
string html2=client.GetString();
这里HttpClient的第三个参数,keepContext设置为真时,HttpClient会自动记录每次交互时服务器对Cookies进行的操 作,同时会以前一次请求的Url为Referer.在这个例子里,获取html2时,会把url1作为Referer, 同时会向服务器传递在获取html1时服务器设置的Cookies. 当然,你也可以在构造HttpClient时直接提供第一次请求要发出的Cookies与Referer:
HttpClient client=new HttpClient(url, new WebContext(cookies, referer), true);
或者,在使用过程中随时修改这些信息:
client.Context.Cookies=cookies;
client.Context.referer=referer;
- 模拟HTML表单提交
HttpClient client=new HttpClient(url);
client.PostingData.Add(fieldName1, filedValue1);
client.PostingData.Add(fieldName2, fieldValue2);
string html=client.GetString();
上面的代码相当于提交了一个有两个input的表单. 在PostingData非空,或者附加了要上传的文件时(请看下面的上传和文件), HttpClient会自动把HttpVerb改成POST, 并将相应的信息附加到Request上.
- 向服务器上传文件
HttpClient client=new HttpClient(url);
client.AttachFile(fileName, fieldName);
client.AttachFile(byteArray, fileName, fieldName);
string html=client.GetString();
这里面的fieldName相当于<input type="file" name="fieldName" />里的fieldName. fileName当然就是你想要上传的文件路径了. 你也可以直接提供一个byte[] 作为文件内容, 但即使如此,你也必须提供一个文件名,以满足HTTP规范的要求.
- 不同的返回形式
字符串: string html = client.GetString();
流: Stream stream = client.GetStream();
字节数组: byte[] data = client.GetBytes();
保存到文件: client.SaveAsFile(fileName);
或者,你也可以直接操作HttpWebResponse: HttpWebResponse res = client.GetResponse();
每调用一次上述任何一个方法,都会导致发出一个HTTP Request, 也就是说,你不能同时得到某个Response的两种返回形式.
另外,调用后它们任意一个之后,你可以通过client.ResponseHeaders来获取服务器返回的HTTP头. - 下载资源的指定部分(用于断点续传,多线程下载)
HttpClient client=new HttpClient(url);
//发出HEAD请求,获取资源长度
int length=client.HeadContentLength();
//只获取后一半内容
client.StartPoint=length/2;
byte[] data=client.GetBytes();
HeadContentLength()只会发出HTTP HEAD请求.根据HTTP协议, HEAD与GET的作用等同, 但是,只返回HTTP头,而不返回资源主体内容. 也就是说,用这个方法,你没法获取一个需要通过POST才能得到的资源的长度,如果你确实有这样的需求,建议你可以通过GetResponse(),然后 从ResponseHeader里获取Content-Length.
计划中还有另外一些功能要加进来,比如断点续传, 多线程下载, 下载进度更新的事件机制等, 正在思考如何与现在的代码融合到一起,期待你的反馈.
注意:使用时应该添加对System.Web.dll的引用,并在使用此类的代码前添加"using System.Web;",不然会无法通过编译.