JAVA敏感詞過濾

一、初始化敏感詞庫

import java.io.BufferedReader;
import java.io.File;
import java.io.FileInputStream;
import java.io.InputStreamReader;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Iterator;
import java.util.Map;
import java.util.Set;

/**
 * 初始化敏感詞庫，將敏感詞加入到HashMap中，構建DFA算法模型
 */
public class SensitiveWordInit {
    private String ENCODING = "utf-8";    //字符編碼
    public HashMap sensitiveWordMap;
    public SensitiveWordInit(){
        super();
    }

    /**
     * 初始化
     */
    public Map initKeyWord(){
        try {
            //讀取敏感詞庫
            Set<String> keyWordSet = readSensitiveWordFile();
            //將敏感詞庫加入到HashMap中
            addSensitiveWordToHashMap(keyWordSet);
            //spring獲取application，然后application.setAttribute("sensitiveWordMap",sensitiveWordMap);
        } catch (Exception e) {
            e.printStackTrace();
        }
        return sensitiveWordMap;
    }

    /**
     * 讀取敏感詞庫，將敏感詞放入HashSet中，構建一個DFA算法模型：<br>
     * 中 = {
     *      isEnd = 0
     *      國 = {<br>
     *           isEnd = 1
     *           人 = {isEnd = 0
     *                民 = {isEnd = 1}
     *                }
     *           男  = {
     *                  isEnd = 0
     *                   人 = {
     *                        isEnd = 1
     *                       }
     *               }
     *           }
     *      }
     *  五 = {
     *      isEnd = 0
     *      星 = {
     *          isEnd = 0
     *          紅 = {
     *              isEnd = 0
     *              旗 = {
     *                   isEnd = 1
     *                  }
     *              }
     *          }
     *      }
     */
    private void addSensitiveWordToHashMap(Set<String> keyWordSet) {
        sensitiveWordMap = new HashMap(keyWordSet.size());     //初始化敏感詞容器，減少擴容操作
        String key = null;
        Map nowMap = null;
        Map<String, String> newWorMap = null;
        //迭代keyWordSet
        Iterator<String> iterator = keyWordSet.iterator();
        while(iterator.hasNext()){
            key = iterator.next();    //關鍵字
            nowMap = sensitiveWordMap;
            for(int i = 0 ; i < key.length() ; i++){
                char keyChar = key.charAt(i);       //轉換成char型
                Object wordMap = nowMap.get(keyChar);       //獲取

                if(wordMap != null){        //如果存在該key，直接賦值
                    nowMap = (Map) wordMap;
                }
                else{     //不存在則，則構建一個map，同時將isEnd設置為0，因為他不是最后一個
                    newWorMap = new HashMap<String,String>();
                    newWorMap.put("isEnd", "0");     //不是最后一個
                    nowMap.put(keyChar, newWorMap);
                    nowMap = newWorMap;
                }

                if(i == key.length() - 1){
                    nowMap.put("isEnd", "1");    //最后一個
                }
            }
        }
    }

    /**
     * 讀取敏感詞庫中的內容，將內容添加到set集合中
     */
    @SuppressWarnings("resource")
    private Set<String> readSensitiveWordFile() throws Exception{
        Set<String> set = null;
        //https://github.com/heqiyoujing/config_file 詞庫地址
        File file = new File("D:\\SensitiveWord.txt");    //讀取文件
        InputStreamReader read = new InputStreamReader(new FileInputStream(file),ENCODING);
        try {
            if(file.isFile() && file.exists()){      //文件流是否存在
                set = new HashSet<String>();
                BufferedReader bufferedReader = new BufferedReader(read);
                String txt = null;
                while((txt = bufferedReader.readLine()) != null){    //讀取文件，將文件內容放入到set中
                    set.add(txt);
                }
            }
            else{         //不存在拋出異常信息
                throw new Exception("敏感詞庫文件不存在");
            }
        } catch (Exception e) {
            throw e;
        }finally{
            read.close();     //關閉文件流
        }
        return set;
    }
}

二、檢查敏感詞並替換

import java.util.HashSet;
import java.util.Iterator;
import java.util.Map;
import java.util.Set;

/**
 * 敏感詞過濾
 */
public class SensitivewordFilter {
    private Map sensitiveWordMap = null;
    public static int minMatchTYpe = 1;      //最小匹配規則
    public static int maxMatchType = 2;      //最大匹配規則
    private static String replaceString = null;
    /**例如：敏感詞中含有中國人、中國
     * 最小匹配規則minMatchTYpe為1時,會匹配出**人，為2時，會匹配出***
     * */
    public static void main(String[] args) throws Exception{
        SensitivewordFilter filter = new SensitivewordFilter();
        System.out.println("敏感詞的數量：" + filter.sensitiveWordMap.size());
        String string = "dfa是面向三級裝配的設計(Design for assembly)的英文簡稱，是指在產品設計階段設計產品使得產品具有良好" +
                "的可裝配性，確保裝配工序簡單、裝配效率高、裝配質量高、裝配不良率低和裝配成本低。面向裝配的設計通過一系" +
                "列有利於裝配的設計指南例如簡化產品設計、減少零件數量等，女女並同裝配工程師一起合作，被逼簡化產品結構，近親使其便於" +
                "裝配，為提高產品質量、縮短產品開發周期和降低產品成本奠定基礎";
        // ------獲取敏感詞---------
        Set<String> set = filter.getSensitiveWord(string, 1);
        System.out.println("含敏感詞的個數為：" + set.size() + "。包含：" + set);
        // ------------------------替換敏感字begin----------------------
        Iterator<String> iterator = set.iterator();
        String word = null;
        while (iterator.hasNext()) {
            word = iterator.next();
            /**
             * 得到word中敏感關鍵詞被替換后的字符串，例如：***
             * */
            getReplaceCharsS("*", word.length());
            /**
             * 將原字符串中的敏感關鍵詞替換成帶有replaceChar
             * 或全部為replaceChar的關鍵詞
             * */
            string = string.replaceAll(word, replaceString);
        }
        // ------------------------替換敏感字end----------------------
        System.out.println(string);
    }

    /**
     * 構造函數，初始化敏感詞庫
     */
    public SensitivewordFilter(){
        sensitiveWordMap = new SensitiveWordInit().initKeyWord();
    }

    /**
     * 判斷文字是否包含敏感字符
     * @param matchType  匹配規則&nbsp;1：最小匹配規則，2：最大匹配規則
     */
    public boolean isContaintSensitiveWord(String txt,int matchType){
        boolean flag = false;
        for(int i = 0 ; i < txt.length() ; i++){
            int matchFlag = this.CheckSensitiveWord(txt, i, matchType); //判斷是否包含敏感字符
            if(matchFlag > 0){    //大於0存在，返回true
                flag = true;
            }
        }
        return flag;
    }

    /**
     * 獲取文字中的敏感詞
     * @param matchType 匹配規則&nbsp;1：最小匹配規則，2：最大匹配規則
     */
    public Set<String> getSensitiveWord(String txt , int matchType){
        Set<String> sensitiveWordList = new HashSet<String>();

        for(int i = 0 ; i < txt.length() ; i++){
            int length = CheckSensitiveWord(txt, i, matchType);    //判斷是否包含敏感字符
            if(length > 0){    //存在,加入list中
                sensitiveWordList.add(txt.substring(i, i+length));
                i = i + length - 1;    //減1的原因，是因為for會自增
            }
        }

        return sensitiveWordList;
    }

    /**
     * 替換敏感字字符,默認*
     */
    public String replaceSensitiveWord(String txt,int matchType,String replaceChar){
        String resultTxt = txt;
        Set<String> set = getSensitiveWord(txt, matchType);     //獲取所有的敏感詞
        Iterator<String> iterator = set.iterator();
        String word = null;
        String replaceString = null;
        while (iterator.hasNext()) {
            word = iterator.next();
            replaceString = getReplaceChars(replaceChar, word.length());
            resultTxt = resultTxt.replaceAll(word, replaceString);
        }

        return resultTxt;
    }

    /**
     * 獲取替換字符串
     */
    private String getReplaceChars(String replaceChar,int length){
        String resultReplace = replaceChar;
        for(int i = 1 ; i < length ; i++){
            resultReplace += replaceChar;
        }

        return resultReplace;
    }

    /**
     * 獲取替換字符串,無返回值
     */
    private static void getReplaceCharsS(String replaceChar,int length){
        replaceString = "";
        String resultReplace = replaceChar;
        for(int i = 1 ; i < length ; i++){
            resultReplace += replaceChar;
        }
        replaceString = resultReplace;
    }

    /**
     * 檢查文字中是否包含敏感字符，檢查規則如下：<br>
     */
    @SuppressWarnings({ "rawtypes"})
    public int CheckSensitiveWord(String txt,int beginIndex,int matchType){
        boolean  flag = false;    //敏感詞結束標識位：用於敏感詞只有1位的情況
        int matchFlag = 0;     //匹配標識數默認為0
        char word = 0;
        Map nowMap = sensitiveWordMap;
        for(int i = beginIndex; i < txt.length() ; i++){
            word = txt.charAt(i);
            nowMap = (Map) nowMap.get(word);     //獲取指定key
            if(nowMap != null){     //存在，則判斷是否為最后一個
                matchFlag++;     //找到相應key，匹配標識+1
                if("1".equals(nowMap.get("isEnd"))){       //如果為最后一個匹配規則,結束循環，返回匹配標識數
                    flag = true;       //結束標志位為true
                    if(SensitivewordFilter.minMatchTYpe == matchType){    //最小規則，直接返回,最大規則還需繼續查找
                        break;
                    }
                }
            }
            else{     //不存在，直接返回
                break;
            }
        }
        if(matchFlag < 2 || !flag){        //長度必須大於等於1，為詞
            matchFlag = 0;
        }
        return matchFlag;
    }

}

三、運行結果