java提取url_java提取网站内部所有URL

展开全部

import java.io.BufferedReader;

import java.io.IOException;

import java.io.InputStreamReader;

import java.net.MalformedURLException;

import java.net.URL;

import java.util.ArrayList;

import java.util.List;

public class GetLinks {

private String webSource;

private String url;

public GetLinks(String url) throws MalformedURLException, IOException {

this.url = Complete(url);

webSource = getWebCon(this.url);

}

private String getWebCon(String strURL) throws MalformedURLException,

IOException {

StringBuffer sb = new StringBuffer();

java.net.URL url = new java.net.URL(strURL);

BufferedReader in = new BufferedReader(new InputStreamReader(url

.openStream()));

String line;

while ((line = in.readLine()) != null) {<