wikiparse.java

来自「可以抓取 wikipeida 中右欄中的人物資訊介紹及其人物圖片。」· Java 代码 · 共 24 行

package parse;
import java.io.FileWriter;
import java.io.IOException;
import java.net.URL;
import java.net.URLEncoder;
import java.util.HashMap;

import org.htmlparser.NodeFilter;
import org.htmlparser.Parser;
import org.htmlparser.Tag;
import org.htmlparser.filters.TagNameFilter;
import org.htmlparser.tags.TableTag;
import org.htmlparser.visitors.NodeVisitor;
public class wikiparse extends NodeVisitor{
	
	  public int count = 0 ;
	  public boolean tr = true;
	  public boolean start =false;
	  public static String name ="";
	  public HashMap<String,String> temp = null;
	  
	  public void visitTag (Tag tag)
	     {
		  NodeFilter filter = new TagNameFilter("TH");    //鍙