A description of needs
Give a url, find out all the links in the url, and supplement it with an absolute path
Two program implementation
/** * * @author Zen Johnny *@date April 29, 2018 at 7:22:44 PM * */ package spider; import java.io.BufferedReader; import java.io.IOException; import java.io.InputStreamReader; import java.io.UnsupportedEncodingException; import java.net.MalformedURLException; import java.net.URL; import java.util.LinkedList; import java.util.List; import java.util.regex.Matcher; import java.util.regex.Pattern; public class WebPageSpider { private static Pattern pattern = null; private static Matcher matches = null; private static BufferedReader br = null; private static StringBuffer text = null; private static URL _url = null; private static List<String> links = null; public static String captureWebPageContent(String url) throws UnsupportedEncodingException, IOException { text = new StringBuffer(); _url = new URL(url); br = new BufferedReader(new InputStreamReader(_url.openStream(), "utf-8")); String line = null; while( (line = br.readLine()) != null) { text.append(line); } return text.toString(); } /* Retrieves and returns all links in a webpage's text @param refUrl: the absolute path of the reference Eg:refAbsoluteUrl -> xxxx.com/X/Y/M/L/test.html?key=35435 */ public static List<String> findLinks(String text, String refAbsoluteUrl) throws MalformedURLException { links = new LinkedList<String>(); String regex = "((href)|(src)){1}=([\"\'])(.*?)\\4";//\\4: If it is preceded by double (single) quotation marks, The end must also be double (single) quotation marks pattern = Pattern.compile(regex); matcher = pattern.matcher (text); String link = null; while(matcher.find()) { link = matcher.group().replaceAll("((href=)|(src=)[\"\'])|([\'\"])", ""); if(link.startsWith(".")) {//If it starts with a relative path, add the base URL by default links.add(revertToAbsolutePath(refAbsoluteUrl, link)); } else if(link.startsWith("/")){//Start with the root path URL tmp_url = new URL(refAbsoluteUrl); links.add(tmp_url.getHost() + link); } else if(link.endsWith("#")){//Use # as the path, that is, the current path (reference path) links.add(refAbsoluteUrl); } else { links.add(link); } } return links; } /* Convert URL path to path list Eg:xxxx.com/X/Y/M/L/test.html?key=35435 => xxxx.com X Y M L test.html?key=35435 */ public static List<String> dirs(String url) { java.util.List<String> dirs = new LinkedList<String>(); String [] dirsArray = url.split("/+"); for(String item : dirsArray) { // System.out.println(item);//test dirs.add(item.trim()); } return dirs; } /* Revert relative path to absolute path @param:curAbsoluteUrl refers to the absolute path @param:relativeSubUrl relative sub path xxxx.com/X/Y/M/L/test.html?key=35435 /hr/ry/ry => xxxx.com/X/Y/M/L/hr/ry/ry xxxx.com/X/Y/M/L/ ./../../G/J/d.x => xxxx.com/X/Y/G/J/d.x */ public static String revertToAbsolutePath(String refAbsoluteUrl, String relativeSubUrl) { List<String> refPaths = dirs(refAbsoluteUrl);//Reference path list List<String> relativePaths = dirs(relativeSubUrl);//relative path list List<String> dirs = new LinkedList<String>(); StringBuffer path = new StringBuffer(); if(refPaths.get(refPaths.size() - 1).matches("(.)*[\\.\\?](.)*")) {//If the last item of the reference path is a file or At the end of the query form, delete the last item // System.out.println(refPaths.get(dirs.size() - 1));//test refPaths.remove(refPaths.size() - 1); } for(String item : relativePaths) { // System.out.println("item:"+item);//test if(item.equals("..")) {//The upper level path, then: delete the last item of refPaths refPaths.remove(refPaths.size() - 1); } else if(!item.equals(".")) {//Not the current path. Or an empty path, that is: a real path if(!item.equals("")) { refPaths.add(item); } else { // System.out.println("【空】");//test } } } for(int item = 0,size = refPaths.size();item<size;item++) {//test path.append(refPaths.get(item) + (item+1 == size?"":"/")); } return path.toString(); } }