几天前使用了Jackson对数据的自定义序列化。突发灵感，利用此方法来简单实现接口返回数据脱敏，故写此文记录。

核心思想是利用Jackson的StdSerializer,@JsonSerialize，以及自己实现的数据脱敏过程。

使用效果如下:

首先在需要进行脱敏的VO字段上面标注相关脱敏注解

调用接口即可看到脱敏效果

实现过程如下:

1. 定义脱敏的过程实现

/**
 * Created by EalenXie on 2021/9/24 15:52
 * 顶级的脱敏器
 */
public interface Desensitization<T> {

    /**
     * 脱敏实现
     *
     * @param target 脱敏对象
     * @return 脱敏返回结果
     */
    T desensitize(T target);

}

比如具体的手机号脱敏器实现

import com.github.Symbol;

import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * Created by EalenXie on 2021/9/24 15:56
 * 手机号脱敏器 默认只保留前3位和后4位
 */
public class PhoneDesensitization implements StringDesensitization {

    /**
     * 手机号正则
     */
    private static final Pattern DEFAULT_PATTERN = Pattern.compile("(13[0-9]|14[579]|15[0-3,5-9]|16[6]|17[0135678]|18[0-9]|19[89])\d{8}");

    /**
     * 手机号脱敏 只保留前3位和后4位
     */
    @Override
    public String desensitize(String target) {
        Matcher matcher = DEFAULT_PATTERN.matcher(target);
        while (matcher.find()) {
            String group = matcher.group();
            target = target.replace(group, group.substring(0, 3) + Symbol.getSymbol(4, Symbol.STAR) + group.substring(7, 11));
        }
        return target;
    }
}

2.定义脱敏注解,并指明了使用的序列化器,注解中声明了使用的脱敏器实现

package com.github.annotation;

import com.fasterxml.jackson.annotation.JacksonAnnotationsInside;
import com.fasterxml.jackson.databind.annotation.JsonSerialize;
import com.github.desensitization.Desensitization;
import com.github.serializer.ObjectDesensitizeSerializer;

import java.lang.annotation.*;

/**
 * Created by EalenXie on 2021/10/8 11:30
 */
@Target({ElementType.FIELD, ElementType.ANNOTATION_TYPE})
@Retention(RetentionPolicy.RUNTIME)
@JacksonAnnotationsInside
@JsonSerialize(using = ObjectDesensitizeSerializer.class)
@Documented
public @interface Desensitize {
    /**
     * 脱敏器实现
     */
    @SuppressWarnings("all")
    Class<? extends Desensitization<?>> desensitization();
}

3. 实现定义的序列化器

import com.fasterxml.jackson.core.JsonGenerator;
import com.fasterxml.jackson.databind.BeanProperty;
import com.fasterxml.jackson.databind.JsonSerializer;
import com.fasterxml.jackson.databind.SerializerProvider;
import com.fasterxml.jackson.databind.ser.ContextualSerializer;
import com.fasterxml.jackson.databind.ser.std.StdSerializer;
import com.github.Symbol;
import com.github.annotation.Desensitize;
import com.github.desensitization.Desensitization;
import com.github.desensitization.DesensitizationFactory;
import com.github.desensitization.StringDesensitization;

import java.io.IOException;

/**
 * Created by EalenXie on 2021/8/9 9:03
 * 脱敏序列化器
 */
public class ObjectDesensitizeSerializer extends StdSerializer<Object> implements ContextualSerializer {

    private transient Desensitization<Object> desensitization;

    protected ObjectDesensitizeSerializer() {
        super(Object.class);
    }

    public Desensitization<Object> getDesensitization() {
        return desensitization;
    }

    public void setDesensitization(Desensitization<Object> desensitization) {
        this.desensitization = desensitization;
    }

    @Override
    public JsonSerializer<Object> createContextual(SerializerProvider prov, BeanProperty property) {
        Desensitize annotation = property.getAnnotation(Desensitize.class);
        return createContextual(annotation.desensitization());
    }

    @SuppressWarnings("unchecked")
    public JsonSerializer<Object> createContextual(Class<? extends Desensitization<?>> clazz) {
        ObjectDesensitizeSerializer serializer = new ObjectDesensitizeSerializer();
        if (clazz != StringDesensitization.class) {
            serializer.setDesensitization((Desensitization<Object>) DesensitizationFactory.getDesensitization(clazz));
        }
        return serializer;
    }

    @Override
    public void serialize(Object value, JsonGenerator gen, SerializerProvider provider) throws IOException {
        Desensitization<Object> objectDesensitization = getDesensitization();
        if (objectDesensitization != null) {
            try {
                gen.writeObject(objectDesensitization.desensitize(value));
            } catch (Exception e) {
                gen.writeObject(value);
            }
        } else if (value instanceof String) {
            gen.writeString(Symbol.getSymbol(((String) value).length(), Symbol.STAR));
        } else {
            gen.writeObject(value);
        }
    }
}

4.代码的设计说明

完整代码可见 : https://github.com/EalenXie/jackson-desensitize

另附 基于Logback的日志脱敏方案(笔者认为这可能是全网最简单快捷的)

原理是利用Logback的自定义日志转换器ClassicConverter

1. 自定义脱敏日志转换器

import ch.qos.logback.classic.pattern.ClassicConverter;
import ch.qos.logback.classic.spi.ILoggingEvent;
import com.github.desensitization.EmailDesensitization;
import com.github.desensitization.IDCardDesensitization;
import com.github.desensitization.PhoneDesensitization;
import com.github.desensitization.StringDesensitization;

import java.util.ArrayList;
import java.util.List;


/**
 * @author EalenXie create on 2021/3/18 10:07
 * 此Converter提供支持日志脱敏
 * 1. 编写此LogbackDesensitizeConverter
 * 2. 正则脱敏 手机号/邮箱/身份证
 */
public class LogbackDesensitizeConverter extends ClassicConverter {

    protected static final List<StringDesensitization> DESENSITIZATION_LIST = new ArrayList<>();

    static {
        // 手机号脱敏
        DESENSITIZATION_LIST.add(new PhoneDesensitization());
        // 邮箱脱敏
        DESENSITIZATION_LIST.add(new EmailDesensitization());
        // 身份证脱敏
        DESENSITIZATION_LIST.add(new IDCardDesensitization());
    }

    @Override
    public String convert(ILoggingEvent event) {
        String content = event.getMessage();
        try {
            for (StringDesensitization desensitization : DESENSITIZATION_LIST) {
                content = desensitization.desensitize(content);
            }
        } catch (Exception e) {
            // ig
        }
        return content;
    }
}

2. 启动类为PatternLayout的静态变量defaultConverterMap新增此自定义转换器

import ch.qos.logback.classic.PatternLayout;
import com.github.filter.LogbackDesensitizeConverter;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.cloud.client.discovery.EnableDiscoveryClient;

/**
 * @author EalenXie create on 2020/11/24 14:16
 */
@EnableDiscoveryClient
@SpringBootApplication
public class ApiGatewayApplication {
    public static void main(String[] args) {
        // 日志处理方案 新增一个Logback的日志脱敏转换器
        PatternLayout.defaultConverterMap.put("m", LogbackDesensitizeConverter.class.getName());
        SpringApplication.run(ApiGatewayApplication.class, args);
    }
}

启动后可以看到日志脱敏效果。

到此这篇关于利用Jackson序列化实现数据脱敏的文章就介绍到这了,更多相关Jackson序列化数据脱敏内容请搜索自由互联以前的文章或继续浏览下面的相关文章希望大家以后多多支持自由互联！

【原URL http://www.yidunidc.com/tw.html复制请保留原URL】