HttpClient抓取网页信息

 import java.io.*;
import org.apache.commons.httpclient.*;
import org.apache.commons.httpclient.methods.GetMethod;
import org.apache.commons.httpclient.params.HttpMethodParams;
public class HttpClientTest{
  public static void main(String[] args) {
  HttpClient httpClient = new HttpClient(); 
  GetMethod getMethod = new GetMethod("http://www.cauyt.edu.cn/showNews.asp?id=4957");
  getMethod.getParams().setParameter(HttpMethodParams.RETRY_HANDLER,
    new DefaultHttpMethodRetryHandler());
  try {
  int statusCode = httpClient.executeMethod(getMethod);
  if (statusCode != HttpStatus.SC_OK) {
    System.err.println("Method failed: "
      + getMethod.getStatusLine());
  } 
  byte[] responseBody = getMethod.getResponseBody();
  FileOutputStream fos=new FileOutputStream("1.htm");
  fos.write(responseBody);
  //System.out.println(new String(responseBody));
  } catch (HttpException e) {
  System.out.println("Please check your provided http address!");
  e.printStackTrace();
  } catch (IOException e) {
  e.printStackTrace();
  } finally { 
  getMethod.releaseConnection();
  }
}
}