asp.net 抓取網頁 自動識別編碼

來源:互聯網
上載者:User

      /// <summary>
        /// 擷取頁面編碼
        /// </summary>
        /// <param name="response">HttpWebResponse</param>
        /// <returns></returns>
        private Encoding GetPageEncoding(HttpWebResponse response)
        {
            //如果發現content-type頭
            string ctype = response.Headers["content-type"];
            string charset = string.Empty;
            if (!string.IsNullOrEmpty(ctype))
            {
                int s = ctype.IndexOf("charset=");

                if (s > -1)
                {
                    charset = ctype.Substring(s+8);//因為“charset=”長度為8位
                }
            }
            //如果沒有發現content-type,只好從指令碼中搜尋了
            if (string.IsNullOrEmpty(charset))
            {
                HttpWebRequest all_codeRequest = (HttpWebRequest)WebRequest.Create(url);
                HttpWebResponse all_codeResponse = (HttpWebResponse)all_codeRequest.GetResponse();

                if (all_codeResponse.StatusCode == HttpStatusCode.OK)
                {
                    StreamReader the_Reader = new StreamReader(all_codeResponse.GetResponseStream(), encoding);
                    ///所有的頁面代碼文本
                    string all_code = the_Reader.ReadToEnd();

                    if (!string.IsNullOrEmpty(all_code))
                    {
                        int s = all_code.IndexOf("charset=");
                        int e = -1;

                        if (s > -1)
                        {
                            s = s + 8;
                            e = all_code.IndexOf("/"", s + 1);

                            if (e > -1)
                            {
                                charset = all_code.Substring(s, e - s);

                                ///去掉開始位置的引號
                                charset = charset.TrimStart(new Char[] { '"' });
                                ///去掉結束位置的引號
                                charset = charset.TrimEnd(new Char[] { '>', '"' });

                            }
                        }

                    }
                    the_Reader.Close();
                    the_Reader.Dispose();
                }
                all_codeResponse.Close();

            }

            if (!string.IsNullOrEmpty(charset))
            {
                try
                {
                    encoding = Encoding.GetEncoding(charset);
                }
                catch (Exception)
                {
                    encoding = Encoding.UTF8;
                }
            }

            return encoding;
        }

        public void SaveFile()
        {
            string all_code = "";
            try
            {

                HttpWebRequest all_codeRequest = (HttpWebRequest)WebRequest.Create(url);

                HttpWebResponse all_codeResponse = (HttpWebResponse)all_codeRequest.GetResponse();

                if (all_codeResponse.StatusCode == HttpStatusCode.OK)
                {

                    encoding = GetPageEncoding(all_codeResponse);


                    StreamReader the_Reader = new StreamReader(all_codeResponse.GetResponseStream(), encoding);

                    all_code = the_Reader.ReadToEnd();

                    FileStream fs = new FileStream("F://test.html", FileMode.Create);

                    StreamWriter sw = new StreamWriter(fs, encoding);

                    sw.WriteLine(all_code);

                    sw.Close();
                    fs.Close();
                    the_Reader.Close();
                    the_Reader.Dispose();
                    all_codeResponse.Close();
                }

            }
            catch (Exception)
            {
                throw;
            }
        }

聯繫我們

該頁面正文內容均來源於網絡整理,並不代表阿里雲官方的觀點,該頁面所提到的產品和服務也與阿里云無關,如果該頁面內容對您造成了困擾,歡迎寫郵件給我們,收到郵件我們將在5個工作日內處理。

如果您發現本社區中有涉嫌抄襲的內容,歡迎發送郵件至: info-contact@alibabacloud.com 進行舉報並提供相關證據,工作人員會在 5 個工作天內聯絡您,一經查實,本站將立刻刪除涉嫌侵權內容。

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.