Java regular expression parsing HTML sample share
- 2020-04-01 02:59:11
- OfStack
package work;
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
import org.apache.commons.httpclient.DefaultHttpMethodRetryHandler;
import org.apache.commons.httpclient.HttpClient;
import org.apache.commons.httpclient.HttpException;
import org.apache.commons.httpclient.HttpStatus;
import org.apache.commons.httpclient.methods.GetMethod;
import org.apache.commons.httpclient.params.HttpMethodParams;
public class chuanboyi {
public static void main(String[] args){
// TODO Auto-generated method stub
StringBuffer html = new StringBuffer();
HttpClient httpclient = new HttpClient();
//Create an instance of the GET method
GetMethod getMethod = new GetMethod("//www.jb51.net");
//Use the default recovery policy provided by the system
getMethod.getParams().setParameter(HttpMethodParams.RETRY_HANDLER, new DefaultHttpMethodRetryHandler());
try{
//Execute the GET method
int statusCode = httpclient.executeMethod(getMethod);
if(statusCode != HttpStatus.SC_OK){
System.out.println("Method is wrong " + getMethod.getStatusLine());
}
InputStream responseBody = getMethod.getResponseBodyAsStream();
BufferedReader reader = new BufferedReader(new InputStreamReader(responseBody,"utf-8"));
String line = reader.readLine();
while(line != null){
html.append(line).append("n");
line = reader.readLine();
}
reader.close();
//Regular expression
String regex = "<form name="compareForm"[\s\S]+>[\s\S]+</form>.*<script.*>";
String regexa ="(?<=<li>)[\s\S]+?(?=</li>)";
Pattern pattern = Pattern.compile(regex);
Matcher m = pattern.matcher(html);
StringBuffer str = new StringBuffer();
int i = 0;
while(m.find()){
str.append(m.group());
}
pattern = Pattern.compile(regexa);
m = pattern.matcher(str);
while(m.find()){
attrs(m.group());
i++;
}
System.out.println(" A total of "+i+" The data! ");
}catch (HttpException e) {
// TODO: handle exception
System.out.println("Please check your provided http address!");
e.printStackTrace();
}catch (IOException e) {
// TODO: handle exception
System.out.println("the line is wrong!");
e.printStackTrace();
}finally{
getMethod.releaseConnection();//Release the link
}
}
public static void attrs(String str){
// To obtain url the Regular expression
String regexURL = "[a-z]+-[0-9]+\.html";
// To obtain Name the Regular expression
String regexName = "(?<=title=")[[\w-\s][^x00-xff]]+(?=")";
// Get the picture Regular expression
String regexPicture = "images.*\.jpg";
Pattern patternURL = Pattern.compile(regexURL);
Pattern patternName = Pattern.compile(regexName);
Pattern patternPicture = Pattern.compile(regexPicture);
Matcher mURL = patternURL.matcher(str);
Matcher mName = patternName.matcher(str);
Matcher mPicture = patternPicture.matcher(str);
if(mName.find()){
System.out.println(" The name :"+mName.group());
}
if(mURL.find()){
System.out.println(" link :"+mURL.group());
}
if(mPicture.find()){
System.out.println(" The picture :"+mPicture.group());
}
}
}