JAVA用URL下載網(wǎng)頁為靜態(tài)頁面...

知識儲備庫_wucl 2011-06-22

展開全文

JAVA用URL下載網(wǎng)頁為靜態(tài)頁面

關(guān)鍵字: java, url, 靜態(tài)頁面

1.可以作為項目的動態(tài)頁面生成靜態(tài)頁面的功能

2.可以當(dāng)做JSP小偷

3.可以當(dāng)做全文網(wǎng)摘

Java代碼

package richin.reflect;
import java.io.BufferedReader;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.InputStreamReader;
import java.net.HttpURLConnection;
import java.net.MalformedURLException;
import java.net.URL;
import java.net.URLConnection;
public class PageStatic {
public static void convert2Html(String sFilePath,String sSavePath,String sHtmlFile) throws IOException
{
PageStatic ru = new PageStatic();
String baseHref="<base href='"+sFilePath+"'></base>";
String filePath = ru.getClass().getResource("/richin/reflect/PageStatic.class").getPath().toString(); //取得項目根目錄
filePath = filePath.substring(1, filePath.indexOf("WEB-INF"));
int HttpResult;
String SavePath = filePath + sSavePath; //保存路徑
URL url=new URL(sFilePath);
URLConnection urlconn=url.openConnection(); //抽象類 URLConnection 是所有類的超類，它代表應(yīng)用程序和 URL 之間的通信鏈接，通過在 URL 上調(diào)用 openConnection 方法創(chuàng)建連接對象
urlconn.addRequestProperty("Accept-Language", "zh-cn");
urlconn.connect(); //使用 connect 方法建立到遠(yuǎn)程對象的實際連接
HttpURLConnection httpconn=(HttpURLConnection)urlconn; //每個 HttpURLConnection 實例都可用于生成單個請求，但是其他實例可以透明地共享連接到 HTTP 服務(wù)器的基礎(chǔ)網(wǎng)絡(luò)
HttpResult=httpconn.getResponseCode(); //getResponseCode可以從 HTTP 響應(yīng)消息獲取狀態(tài)碼
if(HttpResult!=HttpURLConnection.HTTP_OK) {
} else {
String charset=httpconn.getContentType();
//System.out.println(charset);
String charType=charset.substring(charset.lastIndexOf("=")+1);
//System.out.println(charType);
if("text/html".equals(charType))
charType="GBK";
InputStreamReader isr = new InputStreamReader(httpconn.getInputStream(),charType);
BufferedReader in = new BufferedReader(isr);
String inputLine;
if(!SavePath.endsWith("/")) {
SavePath+="/";
}
FileOutputStream fout = new FileOutputStream(SavePath+sHtmlFile);
while ((inputLine = in.readLine()) != null)
{
//System.out.println(inputLine);
if(inputLine.toLowerCase().equals("<head>"))
inputLine=inputLine+System.getProperty("line.separator")+baseHref;//加入basehref
inputLine=inputLine+System.getProperty("line.separator"); //換行
fout.write(inputLine.getBytes());
}
in.close();
fout.close();
}
}
/**
* @param args
* @throws IOException
*/
public static void main(String[] args) throws IOException {
// TODO Auto-generated method stub
convert2Html("http://www.51cto.com/art/200807/79080.htm","/","a4.htm");
}
}