SensitivewordFilter.java

/**
 * DFA敏感词过滤
 * <pre>
 * 五 = { 
 *      isEnd = 0 
 *      星 = { 
 *          isEnd = 0 
 *          红 = { 
 *              isEnd = 0 
 *              旗 = { isEnd = 1 } 
 *          }
 *      }
 * }
 * </pre>
 * @author <a href="mailto:[email protected]">Xiaowei Hu</a>
 * @version 1.0 Aug 6, 2014 10:29:45 AM
 */
public class SensitivewordFilter {
    private static HashMap<Object, Object> sensitivewordMap;
    
    static {
        List<DictKeyWord> listSensitiveword = DictManager.KEY_WORD.getDictList();
        initSensitivewordMap(listSensitiveword);
    }
    
    /**
     * 构建敏感词树
     * 
     * @param wordList
     */
    @SuppressWarnings({ "rawtypes", "unchecked" })
    private static void initSensitivewordMap(List<DictKeyWord> wordList) {
        sensitivewordMap = new HashMap<>(wordList.size() / 2);
        
        Map nowMap = null;
        Map newWorMap = null;
        for (DictKeyWord kw : wordList) {
            nowMap = sensitivewordMap;
            String word = kw.getKeyword();
            
            if (word != null) {
                word = word.trim();
                for (int i = 0; i < word.length(); i++) {

                    char keyChar = word.charAt(i);

                    Object wordMap = nowMap.get(keyChar);
                    
                    if (wordMap != null) {
                        nowMap = (Map) wordMap;
                    } else {
                        newWorMap = new HashMap(2);
                        newWorMap.put("isEnd", "0");
                        nowMap.put(keyChar, newWorMap);
                        nowMap = newWorMap;
                    }

                    if (i == word.length() - 1) {
                        nowMap.put("isEnd", "1");//置结束标志
                    }
                }
            }
        }
    }
    
    /**
     * 判断指定字符是否包含敏感词
     * 
     * @param txt
     * @return
     */
    @SuppressWarnings("rawtypes")
    public static boolean isContaintSensitiveWord(String txt) {
        
        boolean  flag = false;    //敏感词结束标识位：用于敏感词只有1位的情况
        char word = 0;
        Map nowMap = sensitivewordMap;
        a:for (int j = 0; j < txt.length(); j++) {
            for(int i = j; i < txt.length() ; i++){
                word = txt.charAt(i);
                nowMap = (Map) nowMap.get(word);     //获取指定key
                if(nowMap != null){     //存在，则判断是否为最后一个
                    if("1".equals(nowMap.get("isEnd"))){       //如果为最后一个匹配规则,结束循环，返回匹配标识数
                        flag = true;       //结束标志位为true
                        break a;//最小匹配
                    }
                }
                else {
                    nowMap = sensitivewordMap;
                }
            }
        }
        
        return flag;
    }
    
    public static void main(String[] args) {
        System.out.println(isContaintSensitiveWord("小三"));
    }
}