forbidurlfilter.java

java写的搜索引擎网络爬虫 java写的搜索引擎网络爬虫 java写的搜索引擎网络爬虫

/* 
 * FileName ForbidUrlFilter.java
 * Create Time 2006-5-17 16:22:21
 * Author shiwei
 * Descript 判断是否属于禁止范围
 * Version 
 */

package com.snoics.reptile.regex.filter;

import java.util.Iterator;
import java.util.List;
import java.util.regex.Pattern;

import com.snoics.base.util.StringClass;
import com.snoics.base.util.regex.Regex;
import com.snoics.reptile.system.common.Common;
import com.snoics.reptile.system.common.CommonObject;

public class ForbidUrlFilter implements IRegexFilter{
	private String url="";
	private Regex regex=new Regex();
	private CommonObject commonObject=new CommonObject();
	
	public void setUrl(String url) {
		this.url=url;
	}
	
	public String getUrl() {
		return url;
	}
	
	public boolean filter() {
		String regexString=commonObject.getConfigInfo(Common.CONFIGFILE_FORBIDURL);
		List regexStringList=null;
		regexStringList=StringClass.getInterString(Common.STRING_SEPARATE_FLAG,regexString);
		boolean flag=false;
		if((regexStringList!=null)&&(regexStringList.size()>0)){
			Iterator iterator=regexStringList.iterator();
			while(iterator.hasNext()) {
				String regexFlag=(String)iterator.next();
				if(regex.find(url,regexFlag,Pattern.CASE_INSENSITIVE)) {
					return true;
				}
			}
		}
		return flag;
	}
}