Java、C#雙語版HttpHelper類(解決網(wǎng)頁抓取亂碼問題)

在做一些需要抓取網(wǎng)頁的項目時，經(jīng)常性的遇到亂碼問題。最省事的做法是去需要抓取的網(wǎng)站看看具體是什么編碼，然后采用正確的編碼進行解碼就OK了，不過總是一個個頁面親自去判斷也不是個事兒，尤其是你需要大量抓取不同站點的頁面時，比如網(wǎng)頁爬蟲類的程序，這時我們需要做一個相對比較通用的程序，進行頁面編碼的正確識別。

亂碼問題基本上都是編碼不一致導致的，比如網(wǎng)頁編碼使用的是UTF-8，你使用GB2312去讀取，肯定會亂碼。知道了本質(zhì)問題后剩下的就是如何判斷網(wǎng)頁編碼了。GBK、GB2312、UTF-8、BIG-5，一般來說遇到的中文網(wǎng)頁編碼大多是這幾種，簡化下就是只有 GBK和UTF-8兩種，不夸張的說，現(xiàn)在的網(wǎng)站要么是GBK編碼，要么是UTF-8編碼，所以接下來的問題就是判斷站點具體是UTF-8的還是GBK的。

那怎么判斷頁面具體編碼呢？首先查看響應頭的 Content-Type，若響應頭里找不到，再去網(wǎng)頁里查找meta頭，若還是找不到，那沒辦法了，設(shè)置個默認編碼吧，個人推薦設(shè)置成UTF-8。比如訪問博客園首頁http://www.rzrgm.cn/，可以在響應頭里看到 Content-Type: text/html; charset=utf-8，這樣我們就知道博客園是采用utf-8編碼，但并不是所有的網(wǎng)站都會在響應頭Content-Type加上頁面編碼，比如百度的就是Content-Type: text/html，找不到charset，這時只能去網(wǎng)頁里面找<meta http-equiv=Content-Type content="text/html;charset=utf-8">，確認網(wǎng)頁最終編碼，總結(jié)下就是下面幾步

1.響應頭查找Content-Type中的charset，若找到了charset則跳過步驟2，3，直接進行第4步
2.若步驟1得不到charset，則先讀取網(wǎng)頁內(nèi)容，解析meta里面的charset得到頁面編碼
3.若步驟2種還是沒有得到頁面編碼，那沒辦法了設(shè)置默認編碼為UTF-8
4.使用得到的charset重新讀取響應流

通過上面方法基本上能正確解析絕大多數(shù)頁面，實在不能識別的只好親自去核實下具體編碼了

注意：

1.現(xiàn)在站點幾乎都啟用了gzip壓縮支持，所以在請求頭里面加上Accept-Encoding:gzip,deflate，這樣站點會返回壓縮流，能顯著的提高請求效率
2.由于網(wǎng)絡(luò)流不支持流查找操作，也就是只能讀取一次，為了提高效率，所以這里建議將http響應流先讀取到內(nèi)存中，以方便進行二次解碼，沒有必要重新請求去重新獲取響應流

下面分別給出Java和C#版的實現(xiàn)代碼，頁面底部給出了源碼的git鏈接，有需要的童鞋請自行下載

Java實現(xiàn)

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();
        // 若已通過請求頭得到charset，則不需要去html里面繼續(xù)查找
        if (charset == null || charset.equals("")) {
            charset = detectCharset(output.toString());
            // 若在html里面還是未找到charset，則設(shè)置默認編碼為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);

        // 設(shè)置通用的請求屬性
        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());
            }
        }
        // 建立實際的連接
        conn.connect();
        return conn;
    }
}

C#實現(xiàn)

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)
        {
            //用戶https請求
            return true; //總是接受
        }

        public static string SendPost(string url, string data)
        {
            return Send(url, "POST", data, 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 檢測流編碼
                Encoding encoding;

                //檢測響應頭是否返回了編碼類型,若返回了編碼類型則使用返回的編碼
                //注：有時響應頭沒有編碼類型，CharacterSet經(jīng)常設(shè)置為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
                {
                    //若沒有在響應頭找到編碼，則去html找meta頭的charset
                    result = Encoding.Default.GetString(bytes);
                    //在返回的html里使用正則匹配頁面編碼
                    Match match = Regex.Match(result, @"<meta.*charset=""?([\w-]+)""?.*>", RegexOptions.IgnoreCase);
                    if (match.Success)
                    {
                        encoding = Encoding.GetEncoding(match.Groups[1].Value);
                    }
                    else
                    {
                        //若html里面也找不到編碼，默認使用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;
            //有些頁面不設(shè)置用戶代理信息則會抓取不到內(nèi)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"))
            {
                //這里加入解決生產(chǎn)環(huán)境訪問https的問題--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>
        /// 默認(text/html)
        /// </summary>
        public string ContentType { get; set; }

        public string Accept { get; set; }

        public string AcceptEncoding { get; set; }

        /// <summary>
        /// 超時時間(毫秒)默認100000
        /// </summary>
        public int Timeout { get; set; }

        public string UserAgent { get; set; }

        /// <summary>
        /// POST請求時，數(shù)據(jù)是否進行g(shù)zip壓縮
        /// </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";
        }
    }
}

HttpUtil.java

HttpHelper.cs

posted @ 2014-03-02 18:00 懶惰的肥兔閱讀(8828) 評論(17) 收藏舉報

刷新頁面返回頂部

Java、C#雙語版HttpHelper類(解決網(wǎng)頁抓取亂碼問題)

Java實現(xiàn)

C#實現(xiàn)

公告