spring项目中都使用的是Jackson为默认的序列化方式,但是不免有时不满足项目的需要,例如以下场景:
综上各类场景,总而言之就是要对为 null
值的字段进行一些默认值赋值处理,让返回的json中不存在未null的字段,针对此需求无非两种解决方式:一是在业务中手动编码进行各字段赋值(当然部分也需要这么做),二是进行序列化方式配置,让系统在序列化时自动进行赋值
第一种方式的弊端我想大家都能一目了然(重复、烦杂),今天我们主讲第二种实现方式
对比:
使用fastjson的方式配置比较简单,使用也方便,能满足需求,但是本身不够灵活,因为是配置全局生效,要排除的字段需要单独指定序列化方式
使用本次介绍的方式灵活度很高,默认为原本的序列化方式(null 序列化还是 null),不做任何处理,而针对极少场景再使用自定义注解
改变默认序列化方式,让null值序列化为默认值
开发自定义的null值序列化方式,然后配置全局Jackson null值序列化方式进行使用。
其中:
自定义注解
:
主要标识某个类、某个字段需要不为null,需要使用自定义序列化方式,然后再根据注解配置进行自定义的序列化
null值自定义序列化器
:
拦截所有值为null的字段进入自定义序列化器,然后根据自定义注解
配置进行不同的序列化写值
@NeedNotNull
package com.lth.json.jackson;
import java.lang.annotation.ElementType;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
import java.lang.annotation.Target;
/**
*
* json 字段需要不为null
* 只有字段为null时的序列化才会使用此方式,有值的和正常一样
*
*
* @author Tophua
* @since 2021/8/13
*/
@Retention(RetentionPolicy.RUNTIME)
@Target({ElementType.TYPE, ElementType.FIELD})
public @interface NeedNotNull {
/**
* 是否排除,只针对字段
*
* @return true:按原默认处理,false:按此注解处理
*/
boolean isExclude() default false;
/**
* 自定义值,有此值时已此值为主
*
* @return customV
*/
String customV() default "";
/**
* 是否处理Boolean值
*
* @return true:Boolean字段为null时序列化为 boolV 值,false:不处理Boolean字段 null -> null
*/
boolean boolT() default true;
/**
* boolean 默认值
*
* @return boolV
*/
boolean boolV() default false;
/**
* 是否处理Number值
*
* @return true:Number字段为null时序列化为 numberV 值,false:不处理Number字段 null -> null
*/
boolean numberT() default true;
/**
* Number 默认值
*
* @return numberV
*/
int numberV() default 0;
/**
* 是否处理String值
*
* @return true:String字段为null时序列化为 stringV 值,false:不处理String字段 null -> null
*/
boolean stringT() default true;
/**
* String 默认值
*
* @return stringV
*/
String stringV() default "";
/**
* 是否处理Coll(集合和数组)值,null -> 空集合
*
* @return true:Coll字段为null时序列化为 '[]' 值,false:不处理Coll字段 null -> null
*/
boolean collT() default true;
}
自定义null值序列化器
package com.lth.json.jackson;
import cn.hutool.core.collection.ListUtil;
import cn.hutool.core.convert.Convert;
import cn.hutool.core.util.StrUtil;
import com.fasterxml.jackson.core.JsonGenerator;
import com.fasterxml.jackson.databind.BeanProperty;
import com.fasterxml.jackson.databind.JavaType;
import com.fasterxml.jackson.databind.JsonSerializer;
import com.fasterxml.jackson.databind.SerializerProvider;
import lombok.AllArgsConstructor;
import lombok.NoArgsConstructor;
import java.io.IOException;
/**
*
* json 字段为null时的自定义序列化方式
* 只有字段为null时的序列化才会使用此方式,有值的和正常一样
*
*
* @author Tophua
* @since 2021/8/13
*/
@NoArgsConstructor
@AllArgsConstructor
public class NeedNotNullSerialize extends JsonSerializer<Object> {
private BeanProperty property;
@Override
public void serialize(Object value, JsonGenerator gen, SerializerProvider serializers) throws IOException {
if (value != null) {
gen.writeObject(value);
return;
}
JavaType javaType = property.getType();
NeedNotNull needNotNull = property.getAnnotation(NeedNotNull.class);
if (needNotNull == null) {
needNotNull = property.getContextAnnotation(NeedNotNull.class);
}
if (needNotNull == null) {
gen.writeObject(null);
return;
}
// 是否排除
if (needNotNull.isExclude()) {
gen.writeObject(null);
return;
}
// 有自定义值
if (StrUtil.isNotEmpty(needNotNull.customV())) {
gen.writeObject(needNotNull.customV());
return;
}
// bool
if (needNotNull.boolT() && javaType.isTypeOrSubTypeOf(Boolean.class)) {
gen.writeObject(needNotNull.boolV());
return;
}
// Number
if (needNotNull.numberT() && javaType.isTypeOrSubTypeOf(Number.class)) {
gen.writeObject(needNotNull.numberV());
return;
}
// String
if (needNotNull.stringT() && javaType.isTypeOrSubTypeOf(String.class)) {
gen.writeObject(needNotNull.stringV());
return;
}
// 集合、数组
if (needNotNull.collT() && (javaType.isArrayType() || javaType.isCollectionLikeType())) {
gen.writeObject(ListUtil.empty());
return;
}
gen.writeObject(null);
}
}
Jackson 配置
package com.lth.json.config;
import cn.hutool.core.date.DatePattern;
import com.fasterxml.jackson.databind.BeanDescription;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.fasterxml.jackson.databind.SerializationConfig;
import com.fasterxml.jackson.databind.module.SimpleModule;
import com.fasterxml.jackson.databind.ser.BeanPropertyWriter;
import com.fasterxml.jackson.databind.ser.BeanSerializerModifier;
import com.fasterxml.jackson.datatype.jsr310.PackageVersion;
import com.fasterxml.jackson.datatype.jsr310.deser.LocalDateDeserializer;
import com.fasterxml.jackson.datatype.jsr310.deser.LocalDateTimeDeserializer;
import com.fasterxml.jackson.datatype.jsr310.deser.LocalTimeDeserializer;
import com.fasterxml.jackson.datatype.jsr310.ser.LocalDateSerializer;
import com.fasterxml.jackson.datatype.jsr310.ser.LocalDateTimeSerializer;
import com.fasterxml.jackson.datatype.jsr310.ser.LocalTimeSerializer;
import com.lth.json.jackson.NeedNotNullSerialize;
import org.springframework.boot.autoconfigure.condition.ConditionalOnClass;
import org.springframework.context.ApplicationContext;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.http.converter.json.Jackson2ObjectMapperBuilder;
import java.time.LocalDate;
import java.time.LocalDateTime;
import java.time.LocalTime;
import java.time.ZoneId;
import java.time.format.DateTimeFormatter;
import java.util.List;
import java.util.Locale;
import java.util.TimeZone;
/**
* @author skyhua
*/
@Configuration
@ConditionalOnClass(ObjectMapper.class)
public class JacksonConfig {
@Bean
public ObjectMapper objectMapper(ApplicationContext applicationContext) {
Jackson2ObjectMapperBuilder builder = new Jackson2ObjectMapperBuilder();
builder.applicationContext(applicationContext);
builder.locale(Locale.CHINA);
builder.timeZone(TimeZone.getTimeZone(ZoneId.systemDefault()));
builder.simpleDateFormat(DatePattern.NORM_DATETIME_PATTERN);
builder.modules(new JavaTimeModule());
ObjectMapper mapper = builder.createXmlMapper(false).build();
// 为mapper注册一个带有SerializerModifier的Factory,此modifier主要做的事情为:值为null时序列化为默认值
mapper.setSerializerFactory(mapper.getSerializerFactory().withSerializerModifier(new NeedNotNullSerializerModifier()));
return mapper;
}
/**
*
* NeedNotNullSerializerModifier 为bean 设置null值序列化器
*
*
* @author Tophua
* @since 2021/8/14
*/
public static class NeedNotNullSerializerModifier extends BeanSerializerModifier {
@Override
public List<BeanPropertyWriter> changeProperties(SerializationConfig config, BeanDescription beanDesc, List<BeanPropertyWriter> beanProperties) {
beanProperties.forEach(b -> b.assignNullSerializer(new NeedNotNullSerialize(b)));
return beanProperties;
}
}
public static class JavaTimeModule extends SimpleModule {
public JavaTimeModule() {
super(PackageVersion.VERSION);
this.addSerializer(LocalDateTime.class, new LocalDateTimeSerializer(DateTimeFormatter.ofPattern(DatePattern.NORM_DATETIME_PATTERN)));
this.addSerializer(LocalDate.class, new LocalDateSerializer(DateTimeFormatter.ofPattern(DatePattern.NORM_DATE_PATTERN)));
this.addSerializer(LocalTime.class, new LocalTimeSerializer(DateTimeFormatter.ofPattern(DatePattern.NORM_TIME_PATTERN)));
this.addDeserializer(LocalDateTime.class, new LocalDateTimeDeserializer(DateTimeFormatter.ofPattern(DatePattern.NORM_DATETIME_PATTERN)));
this.addDeserializer(LocalDate.class, new LocalDateDeserializer(DateTimeFormatter.ofPattern(DatePattern.NORM_DATE_PATTERN)));
this.addDeserializer(LocalTime.class, new LocalTimeDeserializer(DateTimeFormatter.ofPattern(DatePattern.NORM_TIME_PATTERN)));
}
}
}
默认
@Data
public class Test {
private Object obj;
private Boolean bool;
private Integer intT;
private Long longT;
private BigDecimal decimalT;
private Double doubleT;
private Float floatT;
private String stringV;
private Object[] arrayT;
private List<Integer> collT;
private Set<String> setT;
private String customT;
private Integer customT1;
private LocalDateTime now;
private Test1 test1;
@Data
static class Test1 {
private Integer id;
private String value;
}
}
默认结果
{
"obj": null,
"bool": null,
"intT": null,
"longT": null,
"decimalT": null,
"doubleT": null,
"floatT": null,
"stringV": null,
"arrayT": null,
"collT": null,
"setT": null,
"customT": null,
"customT1": null,
"now": "2021-08-15 17:53:14",
"test1": {
"id": null,
"value": null
}
}
注解用在类上
@Data
@NeedNotNull
public class Test {
private Object obj;
private Boolean bool;
private Integer intT;
private Long longT;
private BigDecimal decimalT;
private Double doubleT;
private Float floatT;
private String stringV;
private Object[] arrayT;
private List<Integer> collT;
private Set<String> setT;
private String customT;
private Integer customT1;
private LocalDateTime now;
private Test1 test1;
@Data
static class Test1 {
private Integer id;
private String value;
}
}
注解在类上结果
{
"obj": null,
"bool": false,
"intT": 0,
"longT": 0,
"decimalT": 0,
"doubleT": 0,
"floatT": 0,
"stringV": "",
"arrayT": [],
"collT": [],
"setT": [],
"customT": "",
"customT1": 0,
"now": "2021-08-15 17:54:00",
"test1": {
"id": null,
"value": null
}
}
可见已有默认值
注解使用在字段上
@Data
public class Test {
private Object obj;
private Boolean bool;
private Integer intT;
private Long longT;
private BigDecimal decimalT;
private Double doubleT;
private Float floatT;
@NeedNotNull
private String stringV;
private Object[] arrayT;
@NeedNotNull
private List<Integer> collT;
private Set<String> setT;
private String customT;
private Integer customT1;
private LocalDateTime now;
private Test1 test1;
@Data
static class Test1 {
private Integer id;
@NeedNotNull
private String value;
}
}
注解用在字段上结果
{
"obj": null,
"bool": null,
"intT": null,
"longT": null,
"decimalT": null,
"doubleT": null,
"floatT": null,
"stringV": "",
"arrayT": null,
"collT": [],
"setT": null,
"customT": null,
"customT1": null,
"now": "2021-08-15 17:58:31",
"test1": {
"id": null,
"value": ""
}
}
花样玩法
@Data
@NeedNotNull
public class Test {
private Object obj;
@NeedNotNull(boolT = false)
private Boolean bool;
private Integer intT;
@NeedNotNull(isExclude = true)
private Long longT;
private BigDecimal decimalT;
@NeedNotNull(customV = "0.5")
private Double doubleT;
private Float floatT;
private String stringV;
@NeedNotNull(customV = "[1,2,3,4]")
private Object[] arrayT;
@NeedNotNull(customV = "[1,2,3,4]")
private List<Integer> collT;
private Set<String> setT;
@NeedNotNull(customV = "花样字符串")
private String customT;
@NeedNotNull(customV = "100")
private Integer customT1;
private LocalDateTime now;
private Test1 test1;
@Data
@NeedNotNull(stringT = false)
static class Test1 {
private Integer id;
private String value;
}
}
花样玩法结果
{
"obj": null,
"bool": null,
"intT": 0,
"longT": null,
"decimalT": 0,
"doubleT": "0.5",
"floatT": 0,
"stringV": "",
"arrayT": "[1,2,3,4]",
"collT": "[1,2,3,4]",
"setT": [],
"customT": "花样字符串",
"customT1": "100",
"now": "2021-08-15 18:02:37",
"test1": {
"id": 0,
"value": null
}
}
各式玩法由各位自行去尝试,也可以再进行扩展
本人意在解决Jackson序列化时对null值的一些自定义序列化方式,让使用者变的简单,使用方式变的灵活,而不是一杆子打死进行全局配置而忽略一些需要null返回值的场景。
此外,个人觉得在spring项目中还是使用Jackson作为序列化方式比较好,因为这是官方默认的方式,fastjson可以使用,但仅限于代码中做某些json转换(毕竟静态调用还是比较香的)。
我是Tophua,欢迎交流
GitHub Gitee