java 自定义json解析注解 复杂json解析 工具类

1.背景

为什么要写这个自定义注解，因为需求需要处理一批比较复杂的json（如果只有一个,直接手动写代码解析就好），众所周知批量且类似的工作，最好抽象出来。这也符合编程的风格，不重复造轮子，但是需要造轮子。【前面铺垫比较长，如果需求比较复杂的json可以直接划到最下面，粘贴自定义注解类 和 自定义注解解析工具类】

2、需求-各式各样的json

一、一星难度json【json对象中不分层】

需要的字段也恰好是对应上的
json

{
"name": "wuyuanshun",
"sex":"男"
}或
[
{
"name": "wuyuanshun",
"sex":"男"
},
{
"name": "liuyuanshun",
"sex":"男"
}
]

java对象

@Data
public class Bean {private String name;private String sex;}

解析方法

public class JsonUtil {public static final ObjectMapper mapper = new ObjectMapper();public static <T> T fromJson(String json, Class<T> clazz) {requireNonNull(json);requireNonNull(clazz);try {return mapper.readValue(json, clazz);} catch (IOException e) {throw new RuntimeException(e);}}/*** json数组转java对象* @param json* @param valueType* @param <T>* @return* @author: wuyuanshn*/public static <T> List<T> jsonArrayToObjectList(String json, Class<T> valueType)  {try {//解析JSONArray redisJsonArray = JSONArray.parseArray(json);List<T> redisList = new ArrayList<>(redisJsonArray.size());//封装for (int i = 0; i < redisJsonArray.size(); i++) {T item = mapper.readValue(redisJsonArray.getString(i), valueType);redisList.add(item);}//返回return redisList;} catch (Exception e) {return null;}}public static void main(String[] args) {String json="xxx";Bean bean = fromJson(json, Bean.class);System.out.println(bean);String jsonList="[xxx,xxx]";List<Bean> beanList = jsonArrayToObjectList(jsonList, Bean.class);System.out.println(beanList);}
}

二、二星难度json【json对象中出现层级】

难度加大一些，比如，出现了层级
json

{
"name": "wuyuanshun",
"sex":"男","like":{"title":"羽毛球","level":1,"time":1672402865000
}
}

简单解析的话，我们可以再新建一个Like对象，如：

@Data
public class Like {private String title;private Integer level;private Long time;
}

然后再bean里加上Like对象即可：

java对象

public class Bean {private String name;private String sex;private Like like;
}

不过需求真的会那么简单，我就不需要写这个文章了。像闯关一下，让我们提升难度，
比如我需要的对象是要同一层级的（如下），要存在一起（比如存数据库表）。当然我们也可以建Like对象再通过代码导入到同一层级。不过我们可以让他简单一些（正题开始了）：
对应解析java对象

@Date
public class Bean {@JsonAnalysisProperty("name")private String name;@JsonAnalysisProperty("sex")private String sex;@JsonAnalysisProperty("like > title")private String likeTitle;@JsonAnalysisProperty("like > level")private Integer likeLevel;@JsonAnalysisProperty("like > time")private Long likeTime;
}

对应解析的bean【自定义注解@JsonAnalysisProperty在文章最下面 目录六】

• name 单层级直接获取
• like > title 多层级 用 > 隔断取下一层级内容 【注意 > 前后有空格】

*测试方法【之后每个测试用测方法】

public static void main(String[] args) {//jsonString json = "{xxxxxxxxxxx}";//自定义对象Bean bean = new Bean();JsonAnalysisPropertyConfig.setObjectByJsonAnalysis(bean,json);System.out.println(bean);}
}

三、三星难度json【json对象中存在数组】

需求取出姓名、性别、爱好名称（like > title）、语文分数。

{"name":"wuyuanshun","sex":"男","like":{"title":"羽毛球","level":1,"time":1672402865000},"examination_results":[{"subject":"数学","date":"2023-01-29","score":98.5},{"subject":"语文","date":"2023-01-29","score":98.5},{"subject":"英语","date":"2023-01-29","score":98.5}]
}

对应解析java对象

@Date
public class Bean {@JsonAnalysisProperty("name")private String name;@JsonAnalysisProperty("sex")private String sex;@JsonAnalysisProperty("like > title")private String likeTitle;@JsonAnalysisProperty("like > level")private Integer likeLevel;@JsonAnalysisProperty("examination_results >> \"subject\":\"语文\" > score")private Double chineseScore;
}

>> 代表之后是数组中的内容，直到"key": "value"这种选择器结束。如果数组到选择器key：value中还有层级，原来的层级( > )需要换成 (>>)，如四星难度json。【注意 >> 前后有空格】

• examination_results >> “subject”:“语文” 找到examination_results数组中key：value为"subject":"语文“的json对象
• > score 继续在json对象层级里找到score的值98.5

四、四星难度json【json对象中数组中存在层级】

{"purchase_crowd":{"interact_data":[{"index_value":{"index_display":"有互动人数","value":{"value":44,"unit":"number"},},"show_list":[{"display":"首购人数占比","value":{"value":0.5909090909090909}}]},{"index_value":{"index_display":"无互动人数"},"show_list":[{"display":"xxx占比","value":{"value":0.02}},{"display":"首购人数占比","value":{"value":0.4444444444444444}}]}]}
}

对应解析java对象

@Date
public class Bean {@JsonAnalysisProperty(defaultValue = "0",value = "purchase_crowd > interact_data >> index_value >> \"index_display\":\"有互动人数\" > value > value") @ApiModelProperty("有互动人数") private String purchaseCrowdInteractDataPeopleNumberInteracting;}

• defaultValue = “0” 是如果json解析中没找到这个字段添加的默认值。
• interact_data >> index_value >> “index_display”:“有互动人数” 这一段都是在数组中选择某一个key：value，来定位自己要找的json对象。找到key：value后，默认在当前层级继续向下选择
• value > value 找到值44。

五、五星难度json【json对象中包含数组、且选择是存在底层级跳转到高层级的】

json同上
如 需求是取index_display 为 "有互动人数"的数组中的 ‘收购人数占比’ value数值、和取index_display 为 "无互动人数"的数组中的‘收购人数占比’value数值

对应解析java对象

@Date
public class Bean {@JsonAnalysisProperty(defaultValue = "0",value = "purchase_crowd > interact_data >> * index_value >> \"index_display\":\"有互动人数\" > show_list >> \"display\":\"首购人数占比\" > value > value") @ApiModelProperty("有互动人数-首购人数占比") private String purchaseCrowdInteractDataPeopleNumberFirstPurchaseRatio;@JsonAnalysisProperty(defaultValue = "0",value = "purchase_crowd > interact_data >> * index_value >> \"index_display\":\"无互动人数\" > show_list >> \"display\":\"首购人数占比\" > value > value") @ApiModelProperty("无互动人数-首购人数占比") private String purchaseCrowdNotInteractDataPeopleNumberFirstPurchaseRatio;
}

自定义注解的逻辑就是像指针一样根据注解中的路由去寻找字段

符号 “ * ” 代表记录指针位置层级，等找到对应的key：value时，返回之前保存的层级。【注意 * 前后有空格】

• interact_data >> * index_value >> “index_display”:“无互动人数” 中的 * 号记录了找到key：value后从 interact_data到 index_value这层开始选择可以继续选择【index_value、show_list】，及通过之前的选择选中了下图中绿色区域。
• > show_list >> “display”:“首购人数占比” 继续从show_list数组中找到对应的key：value（黄色区域）
• > value > value 继续从‘首购人数占比’层级往下寻找 到0.4444…。
  

六、利用自定义注解和配置类解析

 public static void main(String[] args) {//jsonString json = "{xxxxxxxxxxx}";//自定义对象Bean bean = new Bean();JsonAnalysisPropertyConfig.setObjectByJsonAnalysis(bean,json);System.out.println(bean);}
}

1.自定义注解类

package com.wuyuanshun.annotation;import java.lang.annotation.Documented;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
import java.lang.annotation.Target;/*** @program: wys-service* @description: 自定义json解析注解* @author: wuyuanshn* @create: 2022-12-26 17:12**/@Target({java.lang.annotation.ElementType.ANNOTATION_TYPE, java.lang.annotation.ElementType.FIELD, java.lang.annotation.ElementType.METHOD, java.lang.annotation.ElementType.PARAMETER})
@Retention(RetentionPolicy.RUNTIME)
@Documented
public @interface JsonAnalysisProperty {/*** 正常层级选择*/String SPLIT_DEFAULT = " > ";/*** 下层级是数组内元素*/String SPLIT_DEFAULT_ARRAY = " >> ";/*** 数组内 选择返回的层级 默认是最后（最深层）的节点*/String SELECT_ARRAY_DEFAULT_ARRAY = "* ";/*** 核心字段** @return*/String value();/*** 类型 默认0* 1 多层级【后续可以改为枚举类型】** @return*/int type() default 0;/*** 分隔符** @return*/String split() default SPLIT_DEFAULT;/*** 数组内 选择返回的层级 默认是最后（最深层）的节点** @return*/String selectArrayOne() default SELECT_ARRAY_DEFAULT_ARRAY;/*** 标记为数组** @return*/String splitArray() default SPLIT_DEFAULT_ARRAY;/*** 是否忽略** @return*/boolean ignore() default false;/*** 默认值** @return*/String defaultValue() default "";
}

2.自定义注解对应解析工具类

package com.wuyuanshun.annotation;import com.fasterxml.jackson.databind.JsonNode;
import com.fasterxml.jackson.databind.ObjectMapper;
import lombok.extern.slf4j.Slf4j;
import org.springframework.util.StringUtils;import java.lang.reflect.Field;
import java.math.BigDecimal;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Iterator;
import java.util.List;
import java.util.regex.Matcher;
import java.util.regex.Pattern;/*** @program: wys-service* @description: 自定义json解析注解 工具类* @author: wuyuanshn* @create: 2022-12-26 18:38**/
@Slf4j
public class JsonAnalysisPropertyConfig {public static final ObjectMapper mapper = new ObjectMapper();static Pattern GROUP_INDEX_PATTERN = Pattern.compile("\"([\\u4E00-\\u9FA5A-Za-z0-9_]+)\"[ ]*:[ ]*\"([\\u4E00-\\u9FA5A-Za-z0-9_%&',，+!@#^*《》【】\\-（）。;=?$\\x22]+)\"");/*** 根据注解解析 对象中的所有** @param t* @param json* @param <T>*/public static <T> void setObjectByJsonAnalysis(T t, String json) {List<Field> fieldList = getFieldList(t.getClass());for (Field field : fieldList) {setAnalysisProperty(field, t, json);}}/*** 解析对象** @return*/public static List<Field> getFieldList(Class<?> clazz) {List<Field> fieldList = new ArrayList<>();while (clazz != null) {fieldList.addAll(new ArrayList<>(Arrays.asList(clazz.getDeclaredFields())));clazz = clazz.getSuperclass();}Field[] fields = new Field[fieldList.size()];fieldList.toArray(fields);return fieldList;}/*** 根据注解解析 对象中的单个字段** @param field* @param t* @param json* @param <T>*/public static <T> void setAnalysisProperty(Field field, T t, String json) {try {JsonAnalysisProperty annotation = field.getAnnotation(JsonAnalysisProperty.class);if (annotation == null || StringUtils.isEmpty(json)) {return;}//一、 验证参数//忽略boolean ignore = annotation.ignore();String split = annotation.split();String splitArray = annotation.splitArray();String selectArrayOne = annotation.selectArrayOne();String defaultValue = annotation.defaultValue();if (ignore) {return;}field.setAccessible(true);//字段内容Object fieldValue = field.get(t);//有默认值 先这设置默认值 防止后续报错 设置不上(字段没有内容的情况下 再设置默认值)if (!StringUtils.isEmpty(defaultValue) && StringUtils.isEmpty(fieldValue)) {setField(field, t, defaultValue);}//默认-核心字段String codeValue = annotation.value();if (StringUtils.isEmpty(codeValue)) {return;}//二、 处理数据JsonNode jsonNode = mapper.readTree(json);String[] codeValueList = codeValue.split(split);for (int i = 0; i < codeValueList.length; i++) {String codeValueItem = codeValueList[i];if (StringUtils.isEmpty(codeValueItem)) {continue;}//剔除多余空格codeValueItem = codeValueItem.trim();//判断是否是数组 是则处理数组 选择节点JsonNode jsonNodeItem = analysisSplitArrayMax(jsonNode, codeValueItem, splitArray, selectArrayOne);//是选择数组 跳过if (jsonNodeItem != null) {jsonNode = jsonNodeItem;continue;}jsonNode = jsonNode.get(codeValueItem);}String text = jsonNode.asText();//设置值
//            field.set(t, text);setField(field, t, text);//如果为空 且有默认值 设置默认值if (StringUtils.isEmpty(text) && !StringUtils.isEmpty(defaultValue) && StringUtils.isEmpty(fieldValue)) {setField(field, t, defaultValue);}} catch (Exception e) {String error = null;try {StackTraceElement[] stackTrace = e.getStackTrace();StackTraceElement stackTraceElement = stackTrace[0];error = e + "\r\n " + stackTraceElement;} catch (Exception exception) {log.error("JsonAnalysisPropertyConfig setAnalysisProperty exception error {}", e, exception);}log.error("JsonAnalysisPropertyConfig setAnalysisProperty field {}  error {}", field, error);}}/*** 设置字段值 不同类型** @param field* @param t* @param value* @param <T>*/public static <T> void setField(Field field, T t, String value) {try {Object obj = value;Class<?> type = field.getType();if (type.equals(String.class)) {
//                field.set(t, obj);} else if (type.equals(Long.class)) {obj = Long.parseLong(value);} else if (type.equals(Integer.class)) {obj = Integer.parseInt(value);} else if (type.equals(Boolean.class)) {obj = Boolean.parseBoolean(value);} else if (type.equals(BigDecimal.class)) {obj = new BigDecimal(value);} else if (type.equals(Double.class)) {obj = Double.parseDouble(value);} else if (type.equals(Float.class)) {obj = Float.parseFloat(value);}
//            else if (type.equals(Date.class)) {
//                obj = DateUtils.getDate(value);
//            }//其他类型可以在这里添加field.set(t, obj);} catch (Exception e) {log.error("JsonAnalysisPropertyConfig setField 赋值字段失败 field {}; t {}; value {}", field, t, value, e);}}/*** 判断是否是数组,如果是 接着处理** @param jsonNode* @param codeValueItem* @param splitArray*/public static JsonNode analysisSplitArray1(JsonNode jsonNode, String codeValueItem, String splitArray) {//判断是否是数组if (codeValueItem.startsWith(splitArray)) {String key;String value;//查看是否需要选择json数组中的某一个Matcher matcher = GROUP_INDEX_PATTERN.matcher(codeValueItem);if (matcher.find()) {key = matcher.group(1);value = matcher.group(2);} else {return null;}Iterator<JsonNode> elements = jsonNode.elements();//遍历找出对应的数组itemwhile (elements.hasNext()) {JsonNode next = elements.next();String getValue = next.get(key).asText();if (!StringUtils.isEmpty(getValue) && getValue.equals(value)) {return next;}}}return null;}/*** 判断是否是数组,如果是 接着处理** @param jsonNode* @param codeValueItem* @param splitArray*/public static JsonNode analysisSplitArray2(JsonNode jsonNode, String codeValueItem, String splitArray) {//判断是否是数组if (codeValueItem.contains(splitArray)) {String[] keyList = null;keyList = codeValueItem.split(splitArray);//一、平级选择String codeNext = keyList[0];jsonNode = jsonNode.get(codeNext);//二、数组选择Iterator<JsonNode> elements = jsonNode.elements();String key;String value;//查看是否需要选择json数组中的某一个Matcher matcher = GROUP_INDEX_PATTERN.matcher(codeValueItem);if (matcher.find()) {key = matcher.group(1);value = matcher.group(2);} else {//匹配不到筛选key value 但是包含数组 按第一哥个返回（认为数组中只有一个JsonNode 或者取第一个【顺序保证的前提下】）if (elements.hasNext()) {return elements.next();}//取不到数据 认为不是数组return null;}//遍历找出对应的数组itemwhile (elements.hasNext()) {JsonNode next = elements.next();//判断是否需要深层选择if (keyList.length > 2) {for (int i = 0; i < keyList.length; i++) {//跳过最后一个 认为最后一个是key value//跳过第一个 第一个事平级选择if (i == 0 || i == keyList.length - 1) {continue;}next = next.get(keyList[i]);}}String getValue = next.get(key).asText();if (!StringUtils.isEmpty(getValue) && getValue.equals(value)) {return next;}}}return null;}/*** 判断是否是数组,如果是 接着处理** @param jsonNode* @param codeValueItem* @param splitArray* @param selectArrayOne*/public static JsonNode analysisSplitArrayMax(JsonNode jsonNode, String codeValueItem, String splitArray, String selectArrayOne) {//判断是否是数组if (codeValueItem.contains(splitArray)) {String[] keyList = null;keyList = codeValueItem.split(splitArray);//一、层级选择String codeNext = keyList[0];jsonNode = jsonNode.get(codeNext);//二、数组选择Iterator<JsonNode> elements = jsonNode.elements();String key;String value;//查看是否需要选择json数组中的某一个Matcher matcher = GROUP_INDEX_PATTERN.matcher(codeValueItem);if (matcher.find()) {key = matcher.group(1);value = matcher.group(2);} else {//匹配不到筛选key value 但是包含数组 按第一哥个返回（认为数组中只有一个JsonNode 或者取第一个【顺序保证的前提下】）if (elements.hasNext()) {return elements.next();}//取不到数据 认为不是数组return null;}//遍历找出对应的数组itemwhile (elements.hasNext()) {JsonNode next = elements.next();JsonNode returnNext = null;//判断是否需要深层选择if (keyList.length > 2) {for (int i = 0; i < keyList.length; i++) {//跳过最后一个 认为最后一个是key value//跳过第一个 第一个事平级选择if (i == 0 || i == keyList.length - 1) {continue;}String keyItem = keyList[i];if (keyItem.startsWith(selectArrayOne)) {keyItem = keyItem.replace(selectArrayOne, "");returnNext = next;}next = next.get(keyItem);}}String getValue = next.get(key).asText();if (!StringUtils.isEmpty(getValue) && getValue.equals(value)) {//是否选择返回层级if (returnNext != null) {return returnNext;}return next;}}}return null;}public static void main(String[] args) {String a = "aasddd@@\"key_\" :\"value值（asdd，。a）\"";Matcher matcher = GROUP_INDEX_PATTERN.matcher(a);if (matcher.find()) {String key = matcher.group(1);String value = matcher.group(2);System.out.println("key = " + key);System.out.println("value = " + value);/*   key = key_value = value值（asdd，。a）*/}String json = "{\n" +"\"name\": \"wuyuanshun\",\n" +"\"sex\":\"男\",\n" +"\"like\":{\n" +"\t\"title\":\"羽毛球\",\n" +"\t\"level\":1,\n" +"\t\"time\":1672402865000\n" +"},\n" +" \"examination_results\":[\n" +"    {\n" +"\t\"subject\":\"语文\",\n" +"\t\"date\":\"2023-01-29\",\n" +"\t\"score\":98.5\n" +"    }, {\n" +"\t\"subject\":\"数学\",\n" +"\t\"date\":\"2023-01-29\",\n" +"\t\"score\":98.5\n" +"    }, {\n" +"\t\"subject\":\"英语\",\n" +"\t\"date\":\"2023-01-29\",\n" +"\t\"score\":98.5\n" +"    }\n" +"]\n" +"}";Bean bean = new Bean();setObjectByJsonAnalysis(bean,json);System.out.println(bean);}
}

如果业务中json的key出现了 > 这种特殊符合（并且前后带空格的）可以替换。如下：

@JsonAnalysisProperty(defaultValue = "0",value = "purchase_crowd > interact_data >> * index_value >> \"index_display\":\"有互动人数\" > show_list >> \"display\":\"首购人数占比\" > value > value")@ApiModelProperty("成交人群分析-有互动人数-首购人数占比") private String purchaseCrowdInteractDataPeopleNumberFirstPurchaseRatio;
//替换为
@JsonAnalysisProperty(split = " 》 ", splitArray = " 》》 ",  selectArrayOne= "** ", defaultValue = "0",value = "purchase_crowd 》 interact_data 》》 ** index_value 》》 \"index_display\":\"有互动人数\" > show_list 》》 \"display\":\"首购人数占比\" 》 value 》 value")@ApiModelProperty("成交人群分析-有互动人数-首购人数占比") private String purchaseCrowdInteractDataPeopleNumberFirstPurchaseRatio;

七、实战 某个业务场景的一段json

{"purchase_crowd":{"interact_data":[{"index_value":{"index_display":"有互动人数","index_name":"","value":{"value":44,"unit":"number"},"change_value":{"value":0.4943820224719101,"unit":"ratio"}},"show_list":[{"display":"粉丝占比","name":"","value":{"value":0.8636363636363636,"unit":"ratio"}},{"display":"首购人数占比","name":"","value":{"value":0.5909090909090909,"unit":"ratio"}}]},{"index_value":{"index_display":"无互动人数","index_name":"","value":{"value":45,"unit":"number"},"change_value":{"value":0.5056179775280899,"unit":"ratio"},},"show_list":[{"display":"粉丝占比","name":"","value":{"value":0.6888888888888889,"unit":"ratio"}},{"display":"首购人数占比","name":"","value":{"value":0.4444444444444444,"unit":"ratio"}}]}]}
}

对应解析java对象

    @JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> index_value >> \"index_display\":\"有互动人数\" > value > value")@ApiModelProperty("成交人群分析-有互动人数")private String purchaseCrowdInteractDataPeopleNumberInteracting;@JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> index_value >> \"index_display\":\"有互动人数\" > change_value > value")@ApiModelProperty("成交人群分析-有互动人数占比")private String purchaseCrowdInteractDataPeopleNumberRatio;@JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> * index_value >> \"index_display\":\"有互动人数\" > show_list >> \"display\":\"粉丝占比\" > value > value")@ApiModelProperty("成交人群分析-有互动人数-粉丝占比")private String purchaseCrowdInteractDataPeopleNumberFansDataRatio;@JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> * index_value >> \"index_display\":\"有互动人数\" > show_list >> \"display\":\"首购人数占比\" > value > value")@ApiModelProperty("成交人群分析-有互动人数-首购人数占比")private String purchaseCrowdInteractDataPeopleNumberFirstPurchaseRatio;@JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> index_value >> \"index_display\":\"无互动人数\" > value > value")@ApiModelProperty("成交人群分析-无互动人数")private String purchaseCrowdNotInteractDataPeopleNumber;@JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> index_value >> \"index_display\":\"无互动人数\" > change_value > value")@ApiModelProperty("成交人群分析-无互动人数占比")private String purchaseCrowdNotInteractDataPeopleNumberRatio;@JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> * index_value >> \"index_display\":\"无互动人数\" > show_list >> \"display\":\"粉丝占比\" > value > value")@ApiModelProperty("成交人群分析-无互动人数-粉丝占比")private String purchaseCrowdNotInteractDataPeopleNumberFansDataRatio;@JsonAnalysisProperty(defaultValue = "0", value = "purchase_crowd > interact_data >> * index_value >> \"index_display\":\"无互动人数\" > show_list >> \"display\":\"首购人数占比\" > value > value")@ApiModelProperty("成交人群分析-无互动人数-首购人数占比")private String purchaseCrowdNotInteractDataPeopleNumberFirstPurchaseRatio;