Java、C#双语版HttpHelper类(解决网页抓取乱码问题)
在做一些需要抓取网页的项目时，经常性的遇到乱码问题。最省事的做法是去需要抓取的网站看看具体是什么编码，然后采用正确的编码进行解码就OK了，不过总是一个个页面亲自去判断也不是个事儿，尤其是你需要大量抓取不同站点的页面时，比如网页爬虫类的程序，这时我们需要做一个相对比较通用的程序，进行页面编码的正确识别。
乱码问题基本上都是编码不一致导致的，比如网页编码使用的是UTF-8，你使用GB2312去读取，肯定会乱码。知道了本质问题后剩下的就是如何判断网页编码了。GBK、GB2312、UTF-8、BIG-5，一般来说遇到的中文网页编码大多是这几种，简化下就是只有 GBK和UTF-8两种，不夸张的说，现在的网站要么是GBK编码，要么是UTF-8编码，所以接下来的问题就是判断站点具体是UTF-8的还是GBK的。
那怎么判断页面具体编码呢？首先查看响应头的 Content-Type，若响应头里找不到，再去网页里查找meta头，若还是找不到，那没办法了，设置个默认编码吧，个人推荐设置成UTF-8。比如访问博客园首页http://www.cnblogs.com/，可以在响应头里看到 Content-Type: text/html; charset=utf-8，这样我们就知道博客园是采用utf-8编码，但并不是所有的网站都会在响应头Content-Type加上页面编码，比如百度的就是Content-Type: text/html，找不到charset，这时只能去网页里面找<meta http-equiv=Content-Type content="text/html;charset=utf-8">，确认网页最终编码，总结下就是下面几步
- 1.响应头查找Content-Type中的charset，若找到了charset则跳过步骤2，3，直接进行第4步
- 2.若步骤1得不到charset，则先读取网页内容，解析meta里面的charset得到页面编码
- 3.若步骤2种还是没有得到页面编码，那没办法了设置默认编码为UTF-8
- 4.使用得到的charset重新读取响应流
通过上面方法基本上能正确解析绝大多数页面，实在不能识别的只好亲自去核实下具体编码了
注意：
- 1.现在站点几乎都启用了gzip压缩支持，所以在请求头里面加上Accept-Encoding:gzip,deflate，这样站点会返回压缩流，能显著的提高请求效率
- 2.由于网络流不支持流查找操作，也就是只能读取一次，为了提高效率，所以这里建议将http响应流先读取到内存中，以方便进行二次解码，没有必要重新请求去重新获取响应流
下面分别给出Java和C#版的实现代码，页面底部给出了源码的git链接，有需要的童鞋请自行下载
Java实现
package com.cnblogs.lzrabbit.util; import java.io.*; import java.net.*; import java.util.*; import java.util.Map.Entry; import java.util.regex.*; import java.util.zip.*; public class HttpUtil { public static String sendGet(String url) throws Exception { return send(url, "GET", null, null); } public static String sendPost(String url, String param) throws Exception { return send(url, "POST", param, null); } public static String send(String url, String method, String param, Map<String, String> headers) throws Exception { String result = null; HttpURLConnection conn = getConnection(url, method, param, headers); String charset = conn.getHeaderField("Content-Type"); charset = detectCharset(charset); InputStream input = getInputStream(conn); ByteArrayOutputStream output = new ByteArrayOutputStream(); int count; byte[] buffer = new byte[4096]; while ((count = input.read(buffer, 0, buffer.length)) > 0) { output.write(buffer, 0, count); } input.close(); // 若已通过请求头得到charset，则不需要去html里面继续查找 if (charset == null || charset.equals("")) { charset = detectCharset(output.toString()); // 若在html里面还是未找到charset，则设置默认编码为utf-8 if (charset == null || charset.equals("")) { charset = "utf-8"; } } result = output.toString(charset); output.close(); // result = output.toString(charset); // BufferedReader bufferReader = new BufferedReader(new // InputStreamReader(input, charset)); // String line; // while ((line = bufferReader.readLine()) != null) { // if (result == null) // bufferReader.mark(1); // result += line; // } // bufferReader.close(); return result; } private static String detectCharset(String input) { Pattern pattern = Pattern.compile("charset="?([\w\d-]+)"?;?", Pattern.CASE_INSENSITIVE); if (input != null && !input.equals("")) { Matcher matcher = pattern.matcher(input); if (matcher.find()) { return matcher.group(1); } } return null; } private static InputStream getInputStream(HttpURLConnection conn) throws Exception { String ContentEncoding = conn.getHeaderField("Content-Encoding"); if (ContentEncoding != null) { ContentEncoding = ContentEncoding.toLowerCase(); if (ContentEncoding.indexOf("gzip") != 1) return new GZIPInputStream(conn.getInputStream()); else if (ContentEncoding.indexOf("deflate") != 1) return new DeflaterInputStream(conn.getInputStream()); } return conn.getInputStream(); } static HttpURLConnection getConnection(String url, String method, String param, Map<String, String> header) throws Exception { HttpURLConnection conn = (HttpURLConnection) (new URL(url)).openConnection(); conn.setRequestMethod(method); // 设置通用的请求属性 conn.setRequestProperty("Accept", "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8"); conn.setRequestProperty("Connection", "keep-alive"); conn.setRequestProperty("User-Agent", "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36"); conn.setRequestProperty("Accept-Encoding", "gzip,deflate"); String ContentEncoding = null; if (header != null) { for (Entry<String, String> entry : header.entrySet()) { if (entry.getKey().equalsIgnoreCase("Content-Encoding")) ContentEncoding = entry.getValue(); conn.setRequestProperty(entry.getKey(), entry.getValue()); } } if (method == "POST") { conn.setDoOutput(true); conn.setDoInput(true); if (param != null && !param.equals("")) { OutputStream output = conn.getOutputStream(); if (ContentEncoding != null) { if (ContentEncoding.indexOf("gzip") > 0) { output=new GZIPOutputStream(output); } else if(ContentEncoding.indexOf("deflate") > 0) { output=new DeflaterOutputStream(output); } } output.write(param.getBytes()); } } // 建立实际的连接 conn.connect(); return conn; } }
C#实现
using System; using System.Collections; using System.IO; using System.Linq; using System.Net; using System.Net.Security; using System.Security.Cryptography.X509Certificates; using System.Text; using System.Text.RegularExpressions; using System.Web; using System.IO.Compression; using System.Collections.Generic; using System.Collections.Specialized; namespace CSharp.Util.Net { public class HttpHelper { private static bool RemoteCertificateValidate(object sender, X509Certificate certificate, X509Chain chain, SslPolicyErrors errors) { //用户https请求 return true; //总是接受 } public static string SendPost(string url, string data) { return Send(url, "POST", null, null); } public static string SendGet(string url) { return Send(url, "GET", null, null); } public static string Send(string url, string method, string data, HttpConfig config) { if (config == null) config = new HttpConfig(); string result; using (HttpWebResponse response = GetResponse(url, method, data, config)) { Stream stream = response.GetResponseStream(); if (!String.IsNullOrEmpty(response.ContentEncoding)) { if (response.ContentEncoding.Contains("gzip")) { stream = new GZipStream(stream, CompressionMode.Decompress); } else if (response.ContentEncoding.Contains("deflate")) { stream = new DeflateStream(stream, CompressionMode.Decompress); } } byte[] bytes = null; using (MemoryStream ms = new MemoryStream()) { int count; byte[] buffer = new byte[4096]; while ((count = stream.Read(buffer, 0, buffer.Length)) > 0) { ms.Write(buffer, 0, count); } bytes = ms.ToArray(); } #region 检测流编码 Encoding encoding; //检测响应头是否返回了编码类型,若返回了编码类型则使用返回的编码 //注：有时响应头没有编码类型，CharacterSet经常设置为ISO-8859-1 if (!string.IsNullOrEmpty(response.CharacterSet) && response.CharacterSet.ToUpper() != "ISO-8859-1") { encoding = Encoding.GetEncoding(response.CharacterSet == "utf8" ? "utf-8" : response.CharacterSet); } else { //若没有在响应头找到编码，则去html找meta头的charset result = Encoding.Default.GetString(bytes); //在返回的html里使用正则匹配页面编码 Match match = Regex.Match(result, @"<meta.*charset=""?([w-]+)""?.*>", RegexOptions.IgnoreCase); if (match.Success) { encoding = Encoding.GetEncoding(match.Groups[1].Value); } else { //若html里面也找不到编码，默认使用utf-8 encoding = Encoding.GetEncoding(config.CharacterSet); } } #endregion result = encoding.GetString(bytes); } return result; } private static HttpWebResponse GetResponse(string url, string method, string data, HttpConfig config) { HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url); request.Method = method; request.Referer = config.Referer; //有些页面不设置用户代理信息则会抓取不到内容 request.UserAgent = config.UserAgent; request.Timeout = config.Timeout; request.Accept = config.Accept; request.Headers.Set("Accept-Encoding", config.AcceptEncoding); request.ContentType = config.ContentType; request.KeepAlive = config.KeepAlive; if (url.ToLower().StartsWith("https")) { //这里加入解决生产环境访问https的问题--Could not establish trust relationship for the SSL/TLS secure channel ServicePointManager.ServerCertificateValidationCallback = new RemoteCertificateValidationCallback(RemoteCertificateValidate); } if (method.ToUpper() == "POST") { if (!string.IsNullOrEmpty(data)) { byte[] bytes = Encoding.UTF8.GetBytes(data); if (config.GZipCompress) { using (MemoryStream stream = new MemoryStream()) { using (GZipStream gZipStream = new GZipStream(stream, CompressionMode.Compress)) { gZipStream.Write(bytes, 0, bytes.Length); } bytes = stream.ToArray(); } } request.ContentLength = bytes.Length; request.GetRequestStream().Write(bytes, 0, bytes.Length); } else { request.ContentLength = 0; } } return (HttpWebResponse)request.GetResponse(); } } public class HttpConfig { public string Referer { get; set; } /// <summary> /// 默认(text/html) /// </summary> public string ContentType { get; set; } public string Accept { get; set; } public string AcceptEncoding { get; set; } /// <summary> /// 超时时间(毫秒)默认100000 /// </summary> public int Timeout { get; set; } public string UserAgent { get; set; } /// <summary> /// POST请求时，数据是否进行gzip压缩 /// </summary> public bool GZipCompress { get; set; } public bool KeepAlive { get; set; } public string CharacterSet { get; set; } public HttpConfig() { this.Timeout = 100000; this.ContentType = "text/html; charset=" + Encoding.UTF8.WebName; this.UserAgent = "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.117 Safari/537.36"; this.Accept = "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8"; this.AcceptEncoding = "gzip,deflate"; this.GZipCompress = false; this.KeepAlive = true; this.CharacterSet = "UTF-8"; } } }
注：此文章属懒惰的肥兔原创，