c#使用Socket获取网页(含自动跳转, 解压网页)
C#如果访问网页可以用HttpWebRequest, HttpWebResponse两个类.
这儿是用Socket得获取网页的一个例子, 支持301 302跳转. 以及gzip网页解压, 自动切换编码.
public class GetPageWithSocket
{
/// <summary>
/// 使用socket得到网页html
/// </summary>
/// <param name="strUrl">网页url</param>
/// <returns>网页html代码</returns>
public string GetHtml(string strUrl)
{
bool hasRelocated = false;
string strHost = GetHost(strUrl);
string strGetHeader = "GET " + GetRelativeUrl(strUrl) + " HTTP/1.1\r\nHost: " + strHost + "\r\nConnection: Close\r\n\r\n";
byte[] getBytes = Encoding.ASCII.GetBytes(strGetHeader);
try
{
int iTotalCount;
byte[] responseBytes = GetHtmlOriginByte(strHost, getBytes, out iTotalCount); //得到网页
string strResponse = Encoding.UTF8.GetString(responseBytes, 0, iTotalCount);
hasRelocated = HasRelocated(ref strResponse);
if (!hasRelocated) //如果没跳转则转化字符后返回
{
int iStart = FindBodyStartPosition(responseBytes);
//总长度不再计算头http头部了
iTotalCount -= iStart;
int iIndex = strResponse.IndexOf("\r\n\r\n");
string strHeader = strResponse.Substring(0, iIndex);//头部
DecompressWebPage(ref responseBytes, ref iTotalCount, strHeader, iStart);
//转码
strResponse = DecodeWebStringByHttpHeader(responseBytes, iTotalCount, strHeader);
strResponse = DecodeWebStringByHtmlPageInfo(responseBytes, iTotalCount, strResponse);
}
return strResponse;
}
catch (System.Exception ex)
{
return "";
}
}
/// <summary>
/// 得到网页原始字节数组
/// </summary>
/// <param name="strHost">主机头</param>
/// <param name="getBytes">Get字符串的字节数组形式</param>
/// <param name="iTotalCount">接受的字节数</param>
/// <returns>原始网页字节数组</returns>
private byte[] GetHtmlOriginByte(string strHost, byte[] getBytes, out int iTotalCount)
{
int iReponseByteSize = 400 * 1024;
Socket socket = new Socket(AddressFamily.InterNetwork, SocketType.Stream, ProtocolType.Tcp);
socket.Connect(strHost, 80);
socket.Send(getBytes);
byte[] buffer = new byte[256];
byte[] responseBytes = new byte[iReponseByteSize];
int iNumber = socket.Receive(buffer, buffer.Length, SocketFlags.None);
iTotalCount = iNumber;
buffer.CopyTo(responseBytes, 0);
while (iNumber > 0)
{
iNumber = socket.Receive(buffer, buffer.Length, SocketFlags.None);
if (iTotalCount + iNumber >= responseBytes.Length)
{
//重新生成个更大的数组
byte[] temp = new byte[responseBytes.Length * 2];
//原数据copy到新数组中
responseBytes.CopyTo(temp, 0);
buffer.CopyTo(temp, iTotalCount - 1);
responseBytes = temp; //引用变更
}
else
{
buffer.CopyTo(responseBytes, iTotalCount - 1);
}
iTotalCount += iNumber; //索引位置增加
}
return responseBytes;
}
/// <summary>
/// 解压网页
/// </summary>
/// <param name="responseBytes">网页字节数组含http头</param>
/// <param name="iTotalCount">数组长度</param>
/// <param name="strHeader">Http头字符串</param>
/// <param name="iStart">网页正文开始位置</param>
private void DecompressWebPage(ref byte[] responseBytes, ref int iTotalCount, string strHeader, int iStart)
{
byte[] szSwapBuffer = new byte[iTotalCount];
Array.Copy(responseBytes, iStart, szSwapBuffer, 0, iTotalCount);
Regex regZip = new Regex(@"Content-Encoding:\s+gzip[^\n]*\r\n", RegexOptions.IgnoreCase);
if (regZip.IsMatch(strHeader))
{
responseBytes = Decompress(szSwapBuffer);
iTotalCount = responseBytes.Length; //解压后长度变了
}
else
{
responseBytes = szSwapBuffer;
}
}
/// <summary>
/// 根据网页meta标记里面的字符编码解析字符串
/// </summary>
/// <param name="responseBytes">网页内容字节数组(除http头以外的内容)</param>
/// <param name="iTotalCount">网页内容字节数组长度</param>
/// <param name="strResponse">网页内容字符串, 可能已经根据其它转码要求转换过的字符串</param>
/// <returns>转好的字符串</returns>
private string DecodeWebStringByHtmlPageInfo(byte[] responseBytes, int iTotalCount, string strResponse)
{
Regex regGB2312 = new Regex(@"<meta[^>]+Content-Type[^>]+gb2312[^>]*>", RegexOptions.IgnoreCase);
Regex regGBK = new Regex(@"<meta[^>]+Content-Type[^>]+gbk[^>]*>", RegexOptions.IgnoreCase);
Regex regBig5 = new Regex(@"<meta[^>]+Content-Type[^>]+Big5[^>]*>", RegexOptions.IgnoreCase);
if (regGB2312.IsMatch(strResponse) || regGBK.IsMatch(strResponse))
strResponse = Encoding.GetEncoding("GBK").GetString(responseBytes, 0, iTotalCount);
if (regBig5.IsMatch(strResponse))
strResponse = Encoding.GetEncoding("Big5").GetString(responseBytes, 0, iTotalCount);
return strResponse;
}
/// <summary>
/// 根据Http头标记里面的字符编码解析字符串
/// </summary>
/// <param name="responseBytes">网页内容字节数组(除http头以外的内容)</param>
/// <param name="iTotalCount">网页内容字节数组长度</param>
/// <param name="strHeader">http头的字符串</param>
/// <returns>转好的字符串</returns>
private string DecodeWebStringByHttpHeader(byte[] responseBytes, int iTotalCount, string strHeader)
{
string strResponse = "";
if (strHeader.Contains("charset=GBK") || strHeader.Contains("charset=gb2312"))
{
strResponse = Encoding.GetEncoding("GBK").GetString(responseBytes, 0, iTotalCount);
}
else
strResponse = Encoding.UTF8.GetString(responseBytes, 0, iTotalCount);
return strResponse;
}
/// <summary>
/// 是否已跳转
/// </summary>
/// <param name="strResponse">http响应字符串</param>
/// <returns>bool值 是否跳转</returns>
private bool HasRelocated(ref string strResponse)
{
bool hasRelocated = false;
int iIndex = strResponse.IndexOf("\r\n\r\n");
string strHeader = strResponse.Substring(0, iIndex);//头部
if (strResponse.StartsWith("HTTP/1.1 302") || strResponse.StartsWith("HTTP/1.1 301"))//跳转
{
int iStart = strHeader.IndexOf("\r\nLocation: ");
string strRelocateUrl = strHeader.Substring(iStart + "\r\nLocation: ".Length);
int iEnd = strRelocateUrl.IndexOf("\r\n");
strRelocateUrl = strRelocateUrl.Substring(0, iEnd);
strResponse = GetHtml(strRelocateUrl); //递归的获取
hasRelocated = true;
}
return hasRelocated;
}
/// <summary>
/// 查找html开始部分(去除了html头)\r\n\r\n后面第一个byte的位置
/// </summary>
/// <param name="szHtmlBuffer">接受到的byte数组</param>
/// <returns>第一个byte位置</returns>
private int FindBodyStartPosition(byte[] szHtmlBuffer)
{
for (int i = 0; i < szHtmlBuffer.Length - 3; ++i)
{
if (szHtmlBuffer[i] == 13 && szHtmlBuffer[i + 1] == 10 && szHtmlBuffer[i + 2] == 13 && szHtmlBuffer[i + 3] == 10)
return i + 4;
}
return -1;
}
/// <summary>
/// 得到请求资源的相对路径
/// </summary>
/// <param name="strUrl">Url字符串</param>
/// <returns>url的相对路径</returns>
private string GetRelativeUrl(string strUrl)
{
int iIndex = strUrl.IndexOf(@"//");
if (iIndex <= 0)
return "/";
string strTemp = strUrl.Substring(iIndex + 2);
iIndex = strTemp.IndexOf(@"/");
if (iIndex > 0)
return strTemp.Substring(iIndex);
else
return "/";
}
/// <summary>
/// 根据Url得到host
/// </summary>
/// <param name="strUrl">url字符串</param>
/// <returns>host字符串</returns>
private string GetHost(string strUrl)
{
int iIndex = strUrl.IndexOf(@"//");
if (iIndex <= 0)
return "";
string strTemp = strUrl.Substring(iIndex + 2);
iIndex = strTemp.IndexOf(@"/");
if (iIndex > 0)
return strTemp.Substring(0, iIndex);
else
return strTemp;
}
/// <summary>
/// 解压gzip网页
/// </summary>
/// <param name="szSource">压缩过的字符串字节数组</param>
/// <returns>解压后的字节数组</returns>
private byte[] Decompress(byte[] szSource)
{
MemoryStream msSource = new MemoryStream(szSource);
//DeflateStream 也可以这儿
GZipStream stream = new GZipStream(msSource, CompressionMode.Decompress);
byte[] szTotal = new byte[40 * 1024];
long lTotal = 0;
byte[] buffer = new byte[8];
int iCount = 0;
do
{
iCount = stream.Read(buffer, 0, 8);
if (szTotal.Length <= lTotal + iCount) //放大数组
{
byte[] temp = new byte[szTotal.Length * 10];
szTotal.CopyTo(temp, 0);
szTotal = temp;
}
buffer.CopyTo(szTotal, lTotal);
lTotal += iCount;
} while (iCount != 0);
byte[] szDest = new byte[lTotal];
Array.Copy(szTotal, 0, szDest, 0, lTotal);
return szDest;
}
}
class Program
{
public static void Main(string[] args)
{
Console.Write("请输入网址:");
string strUrl = Console.ReadLine();
string strResponse = new GetPageWithSocket().GetHtml(strUrl);
Console.WriteLine(strResponse);
Console.ReadKey();
//File.AppendAllText(@"c:\1.txt", strResponse);
}
}
{
/// <summary>
/// 使用socket得到网页html
/// </summary>
/// <param name="strUrl">网页url</param>
/// <returns>网页html代码</returns>
public string GetHtml(string strUrl)
{
bool hasRelocated = false;
string strHost = GetHost(strUrl);
string strGetHeader = "GET " + GetRelativeUrl(strUrl) + " HTTP/1.1\r\nHost: " + strHost + "\r\nConnection: Close\r\n\r\n";
byte[] getBytes = Encoding.ASCII.GetBytes(strGetHeader);
try
{
int iTotalCount;
byte[] responseBytes = GetHtmlOriginByte(strHost, getBytes, out iTotalCount); //得到网页
string strResponse = Encoding.UTF8.GetString(responseBytes, 0, iTotalCount);
hasRelocated = HasRelocated(ref strResponse);
if (!hasRelocated) //如果没跳转则转化字符后返回
{
int iStart = FindBodyStartPosition(responseBytes);
//总长度不再计算头http头部了
iTotalCount -= iStart;
int iIndex = strResponse.IndexOf("\r\n\r\n");
string strHeader = strResponse.Substring(0, iIndex);//头部
DecompressWebPage(ref responseBytes, ref iTotalCount, strHeader, iStart);
//转码
strResponse = DecodeWebStringByHttpHeader(responseBytes, iTotalCount, strHeader);
strResponse = DecodeWebStringByHtmlPageInfo(responseBytes, iTotalCount, strResponse);
}
return strResponse;
}
catch (System.Exception ex)
{
return "";
}
}
/// <summary>
/// 得到网页原始字节数组
/// </summary>
/// <param name="strHost">主机头</param>
/// <param name="getBytes">Get字符串的字节数组形式</param>
/// <param name="iTotalCount">接受的字节数</param>
/// <returns>原始网页字节数组</returns>
private byte[] GetHtmlOriginByte(string strHost, byte[] getBytes, out int iTotalCount)
{
int iReponseByteSize = 400 * 1024;
Socket socket = new Socket(AddressFamily.InterNetwork, SocketType.Stream, ProtocolType.Tcp);
socket.Connect(strHost, 80);
socket.Send(getBytes);
byte[] buffer = new byte[256];
byte[] responseBytes = new byte[iReponseByteSize];
int iNumber = socket.Receive(buffer, buffer.Length, SocketFlags.None);
iTotalCount = iNumber;
buffer.CopyTo(responseBytes, 0);
while (iNumber > 0)
{
iNumber = socket.Receive(buffer, buffer.Length, SocketFlags.None);
if (iTotalCount + iNumber >= responseBytes.Length)
{
//重新生成个更大的数组
byte[] temp = new byte[responseBytes.Length * 2];
//原数据copy到新数组中
responseBytes.CopyTo(temp, 0);
buffer.CopyTo(temp, iTotalCount - 1);
responseBytes = temp; //引用变更
}
else
{
buffer.CopyTo(responseBytes, iTotalCount - 1);
}
iTotalCount += iNumber; //索引位置增加
}
return responseBytes;
}
/// <summary>
/// 解压网页
/// </summary>
/// <param name="responseBytes">网页字节数组含http头</param>
/// <param name="iTotalCount">数组长度</param>
/// <param name="strHeader">Http头字符串</param>
/// <param name="iStart">网页正文开始位置</param>
private void DecompressWebPage(ref byte[] responseBytes, ref int iTotalCount, string strHeader, int iStart)
{
byte[] szSwapBuffer = new byte[iTotalCount];
Array.Copy(responseBytes, iStart, szSwapBuffer, 0, iTotalCount);
Regex regZip = new Regex(@"Content-Encoding:\s+gzip[^\n]*\r\n", RegexOptions.IgnoreCase);
if (regZip.IsMatch(strHeader))
{
responseBytes = Decompress(szSwapBuffer);
iTotalCount = responseBytes.Length; //解压后长度变了
}
else
{
responseBytes = szSwapBuffer;
}
}
/// <summary>
/// 根据网页meta标记里面的字符编码解析字符串
/// </summary>
/// <param name="responseBytes">网页内容字节数组(除http头以外的内容)</param>
/// <param name="iTotalCount">网页内容字节数组长度</param>
/// <param name="strResponse">网页内容字符串, 可能已经根据其它转码要求转换过的字符串</param>
/// <returns>转好的字符串</returns>
private string DecodeWebStringByHtmlPageInfo(byte[] responseBytes, int iTotalCount, string strResponse)
{
Regex regGB2312 = new Regex(@"<meta[^>]+Content-Type[^>]+gb2312[^>]*>", RegexOptions.IgnoreCase);
Regex regGBK = new Regex(@"<meta[^>]+Content-Type[^>]+gbk[^>]*>", RegexOptions.IgnoreCase);
Regex regBig5 = new Regex(@"<meta[^>]+Content-Type[^>]+Big5[^>]*>", RegexOptions.IgnoreCase);
if (regGB2312.IsMatch(strResponse) || regGBK.IsMatch(strResponse))
strResponse = Encoding.GetEncoding("GBK").GetString(responseBytes, 0, iTotalCount);
if (regBig5.IsMatch(strResponse))
strResponse = Encoding.GetEncoding("Big5").GetString(responseBytes, 0, iTotalCount);
return strResponse;
}
/// <summary>
/// 根据Http头标记里面的字符编码解析字符串
/// </summary>
/// <param name="responseBytes">网页内容字节数组(除http头以外的内容)</param>
/// <param name="iTotalCount">网页内容字节数组长度</param>
/// <param name="strHeader">http头的字符串</param>
/// <returns>转好的字符串</returns>
private string DecodeWebStringByHttpHeader(byte[] responseBytes, int iTotalCount, string strHeader)
{
string strResponse = "";
if (strHeader.Contains("charset=GBK") || strHeader.Contains("charset=gb2312"))
{
strResponse = Encoding.GetEncoding("GBK").GetString(responseBytes, 0, iTotalCount);
}
else
strResponse = Encoding.UTF8.GetString(responseBytes, 0, iTotalCount);
return strResponse;
}
/// <summary>
/// 是否已跳转
/// </summary>
/// <param name="strResponse">http响应字符串</param>
/// <returns>bool值 是否跳转</returns>
private bool HasRelocated(ref string strResponse)
{
bool hasRelocated = false;
int iIndex = strResponse.IndexOf("\r\n\r\n");
string strHeader = strResponse.Substring(0, iIndex);//头部
if (strResponse.StartsWith("HTTP/1.1 302") || strResponse.StartsWith("HTTP/1.1 301"))//跳转
{
int iStart = strHeader.IndexOf("\r\nLocation: ");
string strRelocateUrl = strHeader.Substring(iStart + "\r\nLocation: ".Length);
int iEnd = strRelocateUrl.IndexOf("\r\n");
strRelocateUrl = strRelocateUrl.Substring(0, iEnd);
strResponse = GetHtml(strRelocateUrl); //递归的获取
hasRelocated = true;
}
return hasRelocated;
}
/// <summary>
/// 查找html开始部分(去除了html头)\r\n\r\n后面第一个byte的位置
/// </summary>
/// <param name="szHtmlBuffer">接受到的byte数组</param>
/// <returns>第一个byte位置</returns>
private int FindBodyStartPosition(byte[] szHtmlBuffer)
{
for (int i = 0; i < szHtmlBuffer.Length - 3; ++i)
{
if (szHtmlBuffer[i] == 13 && szHtmlBuffer[i + 1] == 10 && szHtmlBuffer[i + 2] == 13 && szHtmlBuffer[i + 3] == 10)
return i + 4;
}
return -1;
}
/// <summary>
/// 得到请求资源的相对路径
/// </summary>
/// <param name="strUrl">Url字符串</param>
/// <returns>url的相对路径</returns>
private string GetRelativeUrl(string strUrl)
{
int iIndex = strUrl.IndexOf(@"//");
if (iIndex <= 0)
return "/";
string strTemp = strUrl.Substring(iIndex + 2);
iIndex = strTemp.IndexOf(@"/");
if (iIndex > 0)
return strTemp.Substring(iIndex);
else
return "/";
}
/// <summary>
/// 根据Url得到host
/// </summary>
/// <param name="strUrl">url字符串</param>
/// <returns>host字符串</returns>
private string GetHost(string strUrl)
{
int iIndex = strUrl.IndexOf(@"//");
if (iIndex <= 0)
return "";
string strTemp = strUrl.Substring(iIndex + 2);
iIndex = strTemp.IndexOf(@"/");
if (iIndex > 0)
return strTemp.Substring(0, iIndex);
else
return strTemp;
}
/// <summary>
/// 解压gzip网页
/// </summary>
/// <param name="szSource">压缩过的字符串字节数组</param>
/// <returns>解压后的字节数组</returns>
private byte[] Decompress(byte[] szSource)
{
MemoryStream msSource = new MemoryStream(szSource);
//DeflateStream 也可以这儿
GZipStream stream = new GZipStream(msSource, CompressionMode.Decompress);
byte[] szTotal = new byte[40 * 1024];
long lTotal = 0;
byte[] buffer = new byte[8];
int iCount = 0;
do
{
iCount = stream.Read(buffer, 0, 8);
if (szTotal.Length <= lTotal + iCount) //放大数组
{
byte[] temp = new byte[szTotal.Length * 10];
szTotal.CopyTo(temp, 0);
szTotal = temp;
}
buffer.CopyTo(szTotal, lTotal);
lTotal += iCount;
} while (iCount != 0);
byte[] szDest = new byte[lTotal];
Array.Copy(szTotal, 0, szDest, 0, lTotal);
return szDest;
}
}
class Program
{
public static void Main(string[] args)
{
Console.Write("请输入网址:");
string strUrl = Console.ReadLine();
string strResponse = new GetPageWithSocket().GetHtml(strUrl);
Console.WriteLine(strResponse);
Console.ReadKey();
//File.AppendAllText(@"c:\1.txt", strResponse);
}
}
