工具类:
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.HashSet;
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
import java.util.Set;
import org.apache.commons.lang3.StringUtils;
public class WordFilter {
/**
* 存储首字
*/
private static final FilterSet FILTER_SET = new FilterSet();
/**
* 存储节点
*/
private static final Map<Integer, WordNode> WORD_NODES = new HashMap<Integer, WordNode>(1024, 1);
/**
* 停顿词
*/
private static final Set<Integer> STOP_WD_SET = new HashSet<>();
/**
* 敏感词过滤替换
*/
private static final char SIGN = '*';
/**
*
* <p><b>方法描述:</b>初始化,读取敏感词、停顿词数据</p>
* @param words 敏感词
* @return this
*/
public static WordFilter init(List<String> words) {
try {
// 获取敏感词
//addSensitiveWord(readWordFromFile("wordfilter/keywords.txt"));
// 获取停顿词
//addStopWord(readWordFromFile("wordfilter/stopwd.txt"));
//addSensitiveWord(cirBaseFilterWordService.getAll());
//重置数据
WORD_NODES.clear();
FILTER_SET.rest();
STOP_WD_SET.clear();
addSensitiveWord(words);
String stopwd = "! . , # $ % & * ( ) | ? / @ \" ' ; [ ] { } + ~ - _ = ^ < > ! 。 , ¥ ( ) ? 、 “ ‘ ; 【 】 —— …… 《 》";
String[] stopwdArray = stopwd.split(" ");
List<String> stopwdList = new ArrayList<String>();
stopwdList.add(" ");
for (int i = 0; i < stopwdArray.length; i++) {
stopwdList.add(stopwdArray[i]);
}
addStopWord(stopwdList);
} catch (Exception e) {
throw new RuntimeException("初始化过滤器失败");
}
return new WordFilter();
}
/**
* 增加敏感词
* @param path 文件路径
* @return 返回文件内容
*/
@SuppressWarnings("unused")
private static List<String> readWordFromFile(String path) {
List<String> words;
BufferedReader br = null;
try {
br = new BufferedReader(new InputStreamReader(WordFilter.class.getClassLoader().getResourceAsStream(path)));
words = new ArrayList<String>(1200);
for (String buf = ""; (buf = br.readLine()) != null;) {
if (buf == null || buf.trim().equals("")){
continue;
}
words.add(buf);
}
} catch (Exception e) {
throw new RuntimeException(e);
} finally {
try {
if (br != null){
br.close();
}
} catch (IOException e) {
}
}
return words;
}
/**
* 增加停顿词
*
* @param words 停顿词
*/
public static void addStopWord(final List<String> words) {
if (words != null && words.size() > 0) {
char[] chs;
for (String curr : words) {
chs = curr.toCharArray();
for (char c : chs) {
STOP_WD_SET.add(charConvert(c));
}
}
}
}
/**
* 添加DFA节点
* @param words 敏感词
*/
public static void addSensitiveWord(final List<String> words) {
if (words != null && words.size() > 0) {
char[] chs;
int fchar;
int lastIndex;
WordNode fnode; // 首字母节点
for (String curr : words) {
chs = curr.toCharArray();
fchar = charConvert(chs[0]);
if (!FILTER_SET.contains(fchar)) {// 没有首字定义
FILTER_SET.add(fchar); // 首字标志位 可重复add,反正判断了,不重复了
fnode = new WordNode(fchar, chs.length == 1);
WORD_NODES.put(fchar, fnode);
} else {
fnode = WORD_NODES.get(fchar);
if (!fnode.isLast() && chs.length == 1){
fnode.setLast(true);
}
}
lastIndex = chs.length - 1;
for (int i = 1; i < chs.length; i++) {
fnode = fnode.addIfNoExist(charConvert(chs[i]), i == lastIndex);
}
}
}
}
/**
* 过滤判断 将敏感词转化为成屏蔽词
* @param src 内容
* @return 过滤后的内容
*/
public final String doFilter(final String src) {
if (StringUtils.isBlank(src)){
return "";
}
char[] chs = src.toCharArray();
int length = chs.length;
int currc;
int k;
WordNode node;
for (int i = 0; i < length; i++) {
currc = charConvert(chs[i]);
if (!FILTER_SET.contains(currc)) {
continue;
}
node = WORD_NODES.get(currc);
if (node == null){
continue;
}
boolean couldMark = false;
int markNum = -1;
// 单字匹配(日)
if (node.isLast()) {
couldMark = true;
markNum = 0;
}
// 继续匹配(日你/日你妹),以长的优先
// 你-3 妹-4 夫-5
k = i;
for (; ++k < length;) {
int temp = charConvert(chs[k]);
if (STOP_WD_SET.contains(temp)){
continue;
}
node = node.querySub(temp);
if (node == null){
break;
}
if (node.isLast()) {
couldMark = true;
markNum = k - i;
}
}
if (couldMark) {
for (k = 0; k <= markNum; k++) {
chs[k + i] = SIGN;
}
i = i + markNum;
}
}
return new String(chs);
}
/**
* 是否包含敏感词
* @param src 内容
* @return 是否包含敏感词
*/
public final boolean isContains(final String src) {
char[] chs = src.toCharArray();
int length = chs.length;
int currc;
int k;
WordNode node;
for (int i = 0; i < length; i++) {
currc = charConvert(chs[i]);
if (!FILTER_SET.contains(currc)) {
continue;
}
node = WORD_NODES.get(currc);
if (node == null){
continue;
}
boolean couldMark = false;
// 单字匹配(日)
if (node.isLast()) {
couldMark = true;
}
// 继续匹配(日你/日你妹),以长的优先
// 你-3 妹-4 夫-5
k = i;
for (; ++k < length;) {
int temp = charConvert(chs[k]);
if (STOP_WD_SET.contains(temp)){
continue;
}
node = node.querySub(temp);
if (node == null){
break;
}
if (node.isLast()) {
couldMark = true;
}
}
if (couldMark) {
return true;
}
}
return false;
}
/**
* 大写转化为小写 全角转化为半角
*
* @param src 字符
* @return 1
*/
private static int charConvert(char src) {
int r = BCConvert.qj2bj(src);
return (r >= 'A' && r <= 'Z') ? r + 32 : r;
}
}
class WordNode {
/**
* 节点名称
*/
private int value;
/**
* 子节点
*/
private List<WordNode> subNodes;
/**
* 默认false
*/
private boolean isLast;
WordNode(int value) {
this.value = value;
}
WordNode(int value, boolean isLast) {
this.value = value;
this.isLast = isLast;
}
/**
*
* @param subNode subNode
* @return 就是传入的subNode
*/
private WordNode addSubNode(final WordNode subNode) {
if (subNodes == null){
subNodes = new LinkedList<WordNode>();
}
subNodes.add(subNode);
return subNode;
}
/**
* 有就直接返回该子节点, 没有就创建添加并返回该子节点
*
* @param value value
* @param isLast isLast
* @return this
*/
public WordNode addIfNoExist(final int value, final boolean isLast) {
if (subNodes == null) {
return addSubNode(new WordNode(value, isLast));
}
for (WordNode subNode : subNodes) {
if (subNode.value == value) {
if (!subNode.isLast && isLast){
subNode.isLast = true;
}
return subNode;
}
}
return addSubNode(new WordNode(value, isLast));
}
/**
*
* <p><b>方法描述:</b>查询子集</p>
* @param value value
* @return wordnode
*/
public WordNode querySub(final int value) {
if (subNodes == null) {
return null;
}
for (WordNode subNode : subNodes) {
if (subNode.value == value){
return subNode;
}
}
return null;
}
public boolean isLast() {
return isLast;
}
public void setLast(boolean isLast) {
this.isLast = isLast;
}
@Override
public int hashCode() {
return value;
}
}
class FilterSet{
/**
* 元素
*/
private long[] elements;
/**
* 构造函数
*/
public FilterSet() {
elements = new long[1 + (65535 >>> 6)];
}
/**
*
* <p><b>方法描述:</b>添加</p>
* @param no 序号
*/
public void add(final int no) {
elements[no >>> 6] |= (1L << (no & 63));
}
/**
*
* <p><b>方法描述:</b>添加</p>
* @param no 序号
*/
public void add(final int... no) {
for(int currNo : no){
elements[currNo >>> 6] |= (1L << (currNo & 63));
}
}
/**
*
* <p><b>方法描述:</b>添加</p>
* @param no 序号
*/
public void remove(final int no) {
elements[no >>> 6] &= ~(1L << (no & 63));
}
/**
* 重置
*/
public void rest(){
elements = new long[1 + (65535 >>> 6)];
}
/**
*
* @param no 序号
* @return true:添加成功 false:原已包含
*/
public boolean addAndNotify(final int no) {
int eWordNum = no >>> 6;
long oldElements = elements[eWordNum];
elements[eWordNum] |= (1L << (no & 63));
boolean result = elements[eWordNum] != oldElements;
// if (result)
// size++;
return result;
}
/**
*
* @param no 序号
* @return true:移除成功 false:原本就不包含
*/
public boolean removeAndNotify(final int no) {
int eWordNum = no >>> 6;
long oldElements = elements[eWordNum];
elements[eWordNum] &= ~(1L << (no & 63));
boolean result = elements[eWordNum] != oldElements;
return result;
}
/**
*
* <p><b>方法描述:</b>判断是否包含</p>
* @param no 需要
* @return 是否包含
*/
public boolean contains(final int no) {
return (elements[no >>> 6] & (1L << (no & 63))) != 0;
}
/**
*
* <p><b>方法描述:</b>判断是否包含</p>
* @param no 需要
* @return 是否包含
*/
public boolean containsAll(final int... no) {
if (no.length==0) {
return true;
}
for (int currNo : no){
if ((elements[currNo >>> 6] & (1L << (currNo & 63))) == 0) {
return false;
}
}
return true;
}
/**
* 不如直接循环调用contains
* @param no 序号
* @return 是否包含
*/
public boolean containsAllUeslessWay(final int... no) {
long[] elements = new long[this.elements.length];
for (int currNo : no){
elements[currNo >>> 6] |= (1L << (currNo & 63));
}//这一步执行完跟循环调用contains差不多了
for (int i = 0; i < elements.length; i++){
if ((elements[i] & ~this.elements[i]) != 0){
return false;
}
}
return true;
}
/**
* 计算size
* @return 返回元素数量
*/
public int size() {
int size = 0;
for (long element : elements){
size += Long.bitCount(element);
}
return size;
}
}
class BCConvert {
/**
* ASCII表中可见字符从!开始,偏移位值为33(Decimal)
*/
static final char DBC_CHAR_START = 33; // 半角!
/**
* ASCII表中可见字符到~结束,偏移位值为126(Decimal)
*/
static final char DBC_CHAR_END = 126; // 半角~
/**
* 全角对应于ASCII表的可见字符从!开始,偏移值为65281
*/
static final char SBC_CHAR_START = 65281; // 全角!
/**
* 全角对应于ASCII表的可见字符到~结束,偏移值为65374
*/
static final char SBC_CHAR_END = 65374; // 全角~
/**
* ASCII表中除空格外的可见字符与对应的全角字符的相对偏移
*/
static final int CONVERT_STEP = 65248; // 全角半角转换间隔
/**
* 全角空格的值,它没有遵从与ASCII的相对偏移,必须单独处理
*/
static final char SBC_SPACE = 12288; // 全角空格 12288
/**
* 半角空格的值,在ASCII中为32(Decimal)
*/
static final char DBC_SPACE = ' '; // 半角空格
/**
* <PRE>
* 半角字符->全角字符转换
* 只处理空格,!到˜之间的字符,忽略其他
* </PRE>
* @param src 来源内容
* @return 全角内容
*/
public static String bj2qj(String src) {
if (src == null) {
return src;
}
StringBuilder buf = new StringBuilder(src.length());
char[] ca = src.toCharArray();
for (int i = 0; i < ca.length; i++) {
if (ca[i] == DBC_SPACE) { // 如果是半角空格,直接用全角空格替代
buf.append(SBC_SPACE);
} else if ((ca[i] >= DBC_CHAR_START) && (ca[i] <= DBC_CHAR_END)) { // 字符是!到~之间的可见字符
buf.append((char) (ca[i] + CONVERT_STEP));
} else { // 不对空格以及ascii表中其他可见字符之外的字符做任何处理
buf.append(ca[i]);
}
}
return buf.toString();
}
/**
* 半角转换全角
*
* @param src 来源
* @return 全角内容
*/
public static int bj2qj(char src) {
int r = src;
if (src == DBC_SPACE) { // 如果是半角空格,直接用全角空格替代
src = SBC_SPACE;
} else if ((src >= DBC_CHAR_START) && (src <= DBC_CHAR_END)) { // 字符是!到~之间的可见字符
r = src + CONVERT_STEP;
}
return r;
}
/**
* <PRE>
* 全角字符->半角字符转换
* 只处理全角的空格,全角!到全角~之间的字符,忽略其他
* </PRE>
* @param src 来源
* @return 半角内容
*/
public static String qj2bj(String src) {
if (src == null) {
return src;
}
StringBuilder buf = new StringBuilder(src.length());
char[] ca = src.toCharArray();
for (int i = 0; i < src.length(); i++) {
if (ca[i] >= SBC_CHAR_START && ca[i] <= SBC_CHAR_END) { // 如果位于全角!到全角~区间内
buf.append((char) (ca[i] - CONVERT_STEP));
} else if (ca[i] == SBC_SPACE) { // 如果是全角空格
buf.append(DBC_SPACE);
} else { // 不处理全角空格,全角!到全角~区间外的字符
buf.append(ca[i]);
}
}
return buf.toString();
}
/**
* 全角转换半角
*
* @param src 来源
* @return 半角内容
*/
public static int qj2bj(char src) {
int r = src;
if (src >= SBC_CHAR_START && src <= SBC_CHAR_END) { // 如果位于全角!到全角~区间内
r = src - CONVERT_STEP;
} else if (src == SBC_SPACE) { // 如果是全角空格
r = DBC_SPACE;
}
return r;
}
}
Controller 调用:
import javax.annotation.Resource;
import org.springframework.stereotype.Controller;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.ResponseBody;
@Controller
@RequestMapping("/fword")
public class CirCircleController extends BaseController {
@Resource
private ICirBaseFilterWordService cirBaseFilterWordService;
private WordFilter wordFilter = null;
public WordFilter getWordFilter() {
if (wordFilter == null) {
wordFilter = WordFilter.init(cirBaseFilterWordService.getAll());
}
return wordFilter;
}
@ResponseBody
@RequestMapping(value = "/fword")
public ResponseData submitDissolveReason(String words) {
ResponseData responseData = new ResponseData();
if (getWordFilter().isContains(words)) {
return operateFailed("提交内容包含敏感词汇!");
}
responseData.setState(1);
responseData.setMessage("检测正常通过!");
return responseData ;
}
}
查询方法:
import java.util.ArrayList;
import java.util.List;
import javax.annotation.Resource;
import org.springframework.cache.annotation.Cacheable;
import org.springframework.stereotype.Service;
@Service
public class CirBaseFilterWordService extends BaseService<CirBaseFilterWord> implements ICirBaseFilterWordService {
/**
* 敏感词表映射
*/
@Resource
private CirBaseFilterWordMapper cirBaseFilterWordMapper;
@Override
@Cacheable(value="cme-businessarea-entity", key="'FilterWords'")
public List<String> getAll() {
List<CirBaseFilterWord> words = cirBaseFilterWordMapper.selectAll();
List<String> list = new ArrayList<String>();
for (CirBaseFilterWord word : words) {
list.add(word.getWord());
}
return list;
}
}
(注意:cirBaseFilterWordService.getAll()为普通的查询方法,数据库为敏感词库:格式如下:)