网页标题信息采集

--------------------------------------------------------------------------
-----------------------------Cryking原创------------------------------
-----------------------转载请注明出处,谢谢!------------------------

之前做了一个通过IP来扫描网站的,以为能扫到所有的网站,因为IP是能全部遍历的，但是忘记了很多网站是禁止反向解析的。

所以更改了写法,仿照类似爬虫的方法,抓取一个网页的内容,然后遍历其中的网址,然后根据其中的网址再去遍历内容,如此做了个简单的网页标题信息采集工具.

--更新,1.使用多线程处理;2.使用缓存技术减少DB访问;3.优化网页编码获取;4.增加错误日志记录及错误网址检测

主要代码如下:

using System;
using System.Threading;
using System.Collections.Generic;
using System.Windows.Forms;
using System.Linq;
using System.Net;
using System.IO;
using System.Text;
using System.Text.RegularExpressions;
using System.Data;
using System.Data.OracleClient;
using System.Collections;

namespace netCollect
{
    class Program
    {
        static  ulong count = 0;
        static ulong countTimes = 0;
        static public StreamWriter logFile;
        static Hashtable urlHash = new Hashtable();
        static Mutex myMutex = new Mutex();
        static DataTable dt = null;//使用全局dt,将已有的信息读入缓存,以减少数据库访问次数
        static int threadFlag = 0;//线程结束控制标志

        static string GetTitle(string html)
        {


            string regex = @"(?<=<title.*>)([\s\S]*)(?=</title>)";

            //正向预搜索与反向预搜索：http://www.rczjp.cn/HTML/120709/20120409090416.html

            Regex reg = new Regex(regex, RegexOptions.IgnoreCase);

            return reg.Match(html).Value.Trim();

        }

        static MatchCollection GetUrl(string html)
        {
            string regex = @"(?<=http://)[\w\.]+[^/]"; 
            Regex reg = new Regex(regex, RegexOptions.IgnoreCase);
            return reg.Matches(html);

        }

        static void Main(string[] args)
        {
            try
            {
                if (!DBAccess.DBConnect("cry", "cry", "orcl"))
                {
                    MessageBox.Show("数据库连接失败!", "错误001", MessageBoxButtons.OK, MessageBoxIcon.Error);
                    Application.Exit();
                }
                dt = DBAccess.selectDB("select ip,url from net_collect ");//因为ip,url做的联合主键,所以唯一标志一条记录
                Console.WriteLine("请输入开始扫描的网址(2个,逗号隔开):");
                string url = Console.ReadLine();
                string[] urlScan = url.Split(',');//"www.265.com";
                int count = 0;
                logFile = new StreamWriter("d:\\net_collectnew.log", true);
                DateTime startTime = DateTime.Now;
                logFile.WriteLine("开始时间:" + DateTime.Now.ToString());
                Thread t = new Thread(new ParameterizedThreadStart(ipScan));
                t.Start(urlScan[0]);
                Thread t1 = new Thread(new ParameterizedThreadStart(ipScan));
                t1.Start(urlScan[1]);
                while (true) { if (2 == threadFlag) break; };
                DBAccess.DBClose();
                logFile.WriteLine("结束时间:" + DateTime.Now.ToString());
                logFile.Close();
                TimeSpan ts = DateTime.Now - startTime;
                Console.WriteLine("共扫描网址:" + count + "个");
                Console.WriteLine("共扫描次数:" + countTimes + "次");
                Console.WriteLine("总共花费时间:" + ts.ToString());
                Console.ReadKey();
            }
            catch (Exception ex)
            {
                logFile.WriteLine("error:[main]" + ex.Message);
                Console.WriteLine(ex.Message);
            }

        }

        //使用多线程调用
        static void ipScan(object obj)
        {
            string[] html = new string[2];
            html = GetHtmlInfo(obj.ToString(), 60000, Encoding.Default);
            threadFlag++;
            Console.WriteLine(threadFlag+"号线程已停止");
            logFile.WriteLine(threadFlag + "号线程已停止");
        }
        /// <summary>
        /// 获取页面的HTML信息，到标题（</title>）位置结束
        /// </summary>
        /// <param name="url">页面地址</param>
        /// <param name="timeout">超时时间，单位：ms</param>
        /// <param name="EnCodeType">编码</param>
        /// <returns></returns>

        static string[] GetHtmlInfo(string url, int timeout, Encoding EnCodeType)
        {
            string[] urlInfo = new string[2];
            url = url.TrimEnd('"').TrimEnd('?').TrimEnd('\'').TrimEnd(':').TrimEnd('.').TrimEnd('-').Replace("'", "");//干掉网址中不规范的符号
            if (url == "www"||url.Length<3||!url.Replace("www.","").Contains(".")) return null;
            if (!urlHash.ContainsKey(url))
            {
                lock (urlHash)//加锁,防止重复写入
                {
                    myMutex.WaitOne();
                    urlHash.Add(url, url);//hash表来防止重复网址扫描
                    myMutex.ReleaseMutex();
                }
            }
            if (!url.StartsWith("http://") && !url.StartsWith("https://")) { url = "http://" + url; }
            string result = "";
            StreamReader reader = null;
            string temp = "";
            Encoding enc;
            HttpWebRequest request = null;
            HttpWebResponse response = null;
            string encoding = "";
            try
            {

                request = (HttpWebRequest)HttpWebRequest.Create(url);//初始化WebRequest

                request.Timeout = timeout;
                request.UserAgent = "User-Agent:Mozilla/5.0 (compatible; MSIE 6.0; Windows NT 5.2; SV1; .NET CLR 2.0.40607; .NET CLR 1.1.4322; .NET CLR 3.5.30729;.NET CLR 1.0.3705)";
                //request.Headers.Add("Accept-Language", "zh-cn");
                //request.Headers.Add("Accept-Encoding", "gzip, deflate");
                request.Accept = "*/*";
                request.AllowAutoRedirect = false;
                //request.ServicePoint.Expect100Continue = false;
                request.KeepAlive = true;
                //request.CookieContainer = new CookieContainer();

                request.Headers.Add("Accept-Language", "zh-cn,en-us;q=0.5");

                response = (HttpWebResponse)request.GetResponse();//返回来自Internet的响应
                if (response.StatusCode == System.Net.HttpStatusCode.MovedPermanently)//获取重定向的网页
                {
                    request = (HttpWebRequest)HttpWebRequest.Create(response.Headers["Location"]);//初始化WebRequest
                    response = (HttpWebResponse)request.GetResponse();//返回来自Internet的响应
                }
                System.Net.IPAddress[] addresslist = Dns.GetHostEntry(url.Replace("https://","").Replace("http://","")).AddressList;
                if (addresslist.Length != 0) urlInfo[0] = addresslist[0].ToString().Trim();//获取IP
                else { logFile.WriteLine("error:获取IP失败! [URL]=" + url); return null; }
                countTimes++;
                Console.WriteLine("第[" + countTimes + "]次Scan, url=" + url.Replace("https://", "").Replace("http://", ""));
                //DataTable dt=DBAccess.selectDB("select count(*) from net_collect where ip='"+urlInfo[0].ToString()+"' and url='"+url+"' ");
                DataRow[] dr = dt.Select(@"ip='" + urlInfo[0].ToString() + "' and url='" + url.TrimEnd('"').TrimEnd('?').TrimEnd('\'') + "' ");
                if (dr.Length>0&&count!=0)//递归终止条件
                    return null;
                DataTable dt1 = DBAccess.selectDB("select count(*) from net_collect where ip='" + urlInfo[0].ToString() + "' and url='" + url + "' ");
                if (null == dt1) { logFile.WriteLine("error:数据库查询net_collect失败! [URL]=" + url); return null; }
                if (dt1.Rows[0][0].ToString() != "0" && count != 0)//递归终止条件
                    return null;
                lock (url)//加锁,防止重复写入
                { myMutex.WaitOne(); encoding = GetHtmlCode(url); myMutex.ReleaseMutex(); }
                if (encoding == null || encoding == string.Empty || encoding.Contains("--"))
                {
                    encoding = response.CharacterSet;
                    encoding = (encoding == "ISO-8859-1") ? "gb2312" : encoding;
                }
                enc = Encoding.GetEncoding(encoding);
                if (response.StatusCode == System.Net.HttpStatusCode.OK)
                {

                    StringBuilder builder = new StringBuilder();
                    Stream stream = response.GetResponseStream();

                    reader = new StreamReader(stream, enc);
                    string tmp = "";
                    while ((temp = reader.ReadLine()) != null)
                    {
                        builder.Append(temp);

                        tmp = builder.ToString();

                        if (urlInfo[1]==null&&tmp.IndexOf("</title>") > 0 )
                            urlInfo[1] = GetTitle(tmp);

                        builder.Append("\r\n");


                    }
                    
                    result = builder.ToString();
                    
                }
                if (urlInfo[1].Length > 3000) urlInfo[1] = urlInfo[1].Substring(0, 2800);
                    MatchCollection mc = GetUrl(result);

                    if (urlInfo[0] != null && urlInfo[1] != "无法连接到远程服务器" && url!=null)
                    {

                        DBAccess.DBExecSql(@"insert into net_collect values('" + urlInfo[0] + "',default,q'[" + urlInfo[1] + "]','" + url + "',default)");
                        count++;
                        Console.WriteLine("[" + count + "]" + "已写入数据库,信息为[url:]" + url + "  [ip:]" + urlInfo[0]);
                        logFile.WriteLine("[" + count + "]" + "已写入数据库,信息为[url:]" + url + "  [ip:]" + urlInfo[0]);
                    }
                    
                    if(mc.Count!=0)
                    foreach (Match match in mc)
                    {
                        if (urlHash.ContainsKey(match.ToString())) continue;
                        if (match.ToString().TrimEnd('"').TrimEnd('?').TrimEnd('\'') == "www") continue;
                        Console.WriteLine("开始扫描网址:" + match.ToString());
                        logFile.WriteLine("开始扫描网址:" + match.ToString());
                        GetHtmlInfo(match.ToString().TrimEnd('"').TrimEnd('?').TrimEnd('\''), timeout, EnCodeType);//递归
                        
                    }

                    

                    return urlInfo;

            }

            catch (Exception ex)
            {
                logFile.WriteLine("error:--"+ex.Message+" [URL]="+url);
                Console.WriteLine(ex.Message);
                urlInfo[1] = ex.Message;
                return urlInfo;
            }

            finally { if (reader != null) { reader.Close(); } if (response != null) { response.Close(); } if (request != null) { request.Abort(); } }

        }

        static string GetHtmlCode(string url)
        {
            string charset = "";
            try
            {
                string htmlCode;
                string temp = "";
                string pattern = @"(?i)\bcharset=(?<charset>[-a-zA-Z_0-9]+)";
                HttpWebRequest webRequest = (System.Net.HttpWebRequest)System.Net.WebRequest.Create(url);
                webRequest.Timeout = 30000;
                webRequest.Method = "GET";
                webRequest.UserAgent = "User-Agent:Mozilla/5.0 (compatible; MSIE 6.0; Windows NT 5.2; SV1; .NET CLR 2.0.40607; .NET CLR 1.1.4322; .NET CLR 3.5.30729;.NET CLR 1.0.3705)";
                webRequest.Headers.Add("Accept-Encoding", "gzip, deflate");
                HttpWebResponse webResponse = (System.Net.HttpWebResponse)webRequest.GetResponse();
                if (webResponse.ContentEncoding.ToLower() == "gzip")//如果使用了GZip则先解压
                {
                    using (System.IO.Stream streamReceive = webResponse.GetResponseStream())
                    {
                        using (var zipStream =
                            new System.IO.Compression.GZipStream(streamReceive, System.IO.Compression.CompressionMode.Decompress))
                        {
                            using (StreamReader sr = new System.IO.StreamReader(zipStream, Encoding.Default))
                            {
                                string tmp = "";
                                while ((temp = sr.ReadLine()) != null)
                                {
                                    StringBuilder builder = new StringBuilder();
                                    builder.Append(temp);

                                    tmp = builder.ToString();

                                    if (tmp.IndexOf("charset") > 0)
                                    {
                                        charset = Regex.Match(tmp, pattern).Groups["charset"].Value;
                                        if (charset.Length < 3) //charset = Regex.Match(tmp, "/<meta(.*)charset=(\'|\")(.*)(\'|\")>/i").Groups["charset"].Value;
                                        { //匹配 <meta charset="utf-8" /> 这种情况
                                            charset = tmp.Substring(tmp.IndexOf("=") + 1, tmp.Length - tmp.IndexOf("=") - 1).Replace("\"", "").Replace("/>", "").Trim();
                                            charset = charset.ToLower().Contains("utf-8") ? "utf-8" : "gb2312";//强制2选1,呵呵
                                        }
                                        return charset;
                                    }
                                    builder.Append("\r\n");
                                }
                            }
                        }
                    }
                }
                else
                {
                    using (System.IO.Stream streamReceive = webResponse.GetResponseStream())
                    {
                        using (System.IO.StreamReader sr = new System.IO.StreamReader(streamReceive, Encoding.Default))
                        {
                            string tmp = "";
                            while ((temp = sr.ReadLine()) != null)
                            {
                                StringBuilder builder = new StringBuilder();
                                builder.Append(temp);

                                tmp = builder.ToString();

                                if (tmp.IndexOf("charset") > 0)
                                {
                                    charset = Regex.Match(tmp, pattern).Groups["charset"].Value;
                                    if (charset.Length<3) //charset = Regex.Match(tmp, "/<meta(.*)charset=(\'|\")(.*)(\'|\")>/i").Groups["charset"].Value;
                                    { //匹配 <meta charset="utf-8" /> 这种情况
                                        charset = tmp.Substring(tmp.IndexOf("=") + 1, tmp.Length - tmp.IndexOf("=") - 1).Replace("\"", "").Replace("/>", "").Trim();
                                        charset = charset.ToLower().Contains("utf-8") ? "utf-8" : "gb2312";//强制2选1,呵呵
                                    }
                                    return charset;
                                }
                                builder.Append("\r\n");
                             }

                        }
                    }
                }
                webResponse.Close();
                webRequest.Abort();
                return charset;
            }
            catch (Exception ex)
            {
                logFile.WriteLine("error:--" + ex.Message + " [charset]=" + charset);
                Console.WriteLine(ex.Message);
                return null;
            }
        }
    }
}

运行如图:

---日志文件部分内容如下:

开始时间:2013-5-14 19:22:37
[1]已写入数据库,信息为[url:]http://www.265.com [ip:]203.208.37.17
开始扫描网址:www.w3.org
开始扫描网址:www.google.com.hk
开始扫描网址:news.google.com.hk
开始扫描网址:ditu.google.cn
[2]已写入数据库,信息为[url:]http://ditu.google.cn [ip:]203.208.36.18
开始扫描网址:schema.org
[3]已写入数据库,信息为[url:]http://schema.org [ip:]74.125.31.100
开始扫描网址:www.google.com
开始扫描网址:blog.schema.org"
开始扫描网址:www.google.cn
[4]已写入数据库,信息为[url:]http://www.google.cn [ip:]203.208.36.19
开始扫描网址:translate.google.cn
[5]已写入数据库,信息为[url:]http://translate.google.cn [ip:]203.208.37.17
开始扫描网址:picasaweb.google.com
开始扫描网址:video.google.cn
[6]已写入数据库,信息为[url:]http://video.google.cn [ip:]203.208.46.177
开始扫描网址:www.miibeian.gov.cn
开始扫描网址:www.gstatic.com
开始扫描网址:translate.google.com

...

-----------------------------------------

不上传工具了,如对此感兴趣者,可直接联系我索要源码及工具.QQ:278676125