【Gson】【3】实例演习
【本文范围】:
本文并非JSON知识讲解资料,亦非GSON知识讲解资料,而是通过实例让开发人员了解通过Gson如何使Java对象和json对象进行相互转换。
【JSON参考资料】:
Json快速入门:Http://www.w3school.com.cn/json/index.asp
Json官网:http://www.json.org/
【GSON参考资料】:
Gson官网:http://code.Google.com/p/google-gson/
一、JSON基础知识:
1、Json概念:
Json全称javascript object notation(Javascript对象表示法),是轻量级的文本数据交换格式,它独立于语言和平台,用来描述数据对象,具有自我描述性、容易理解等特性。
2、举例:
{“TeamName":"hr", "NewProject":true, "Leader":"wang", "Employee":[{"name":"li", "BelongTo":"train"}, {"name":"kpi", "BelongTo":"Security"}, {"name":"liu", "BelonTo":"service"}]}
这个Json对象描述了hr项目组。项目组名称为hr,是一个新项目,Leader是wang,有三个员工,分别为负责培训的li,负责kpi的zheng和负责服务的liu
3、Json语法
数据以健值对形式描述(如:"TeamName":"hr");
数据以逗号分隔;
花括号保存对象(如{"Name":"li", "BelongTo":"train"});
方括号保存数组(如"csn":[20,21,22,23]);
4、Json值:
数字(整数或浮点数);
字符串(在双引号中);
逻辑值(true或false);
数组(在方括号中);
对象(在花括号中);
null
二、实例演习:
实例1:在查询告警窗口中对告警流水号为2012的记录做“清除”操作
【思路】:
假如清除操作的命令码为41,那么向服务端请求的JSON对象应该为{“cmdCode”:41, “csn”:2012}
【演示代码】:
1、定义AlarmClearCondition.java
public class AlarmClearCondition
{
private int cmdCode = 0;
private int csn = 0;
public void setCmdCode(int cmdCode)
{
thisNaNdCode = cmdCode;
}
public void setCsn(int csn)
{
this.csn= csn;
}
}
2、使用Gson把AlarmClearCondition对象转换为Json对象
AlarmClearConditioncondition = new AlarmClearCondition();
condition.setCmdCode(41);
condition.setCsn(2012);
Gson gson = newGson();
String jsonCond= gson.toJson(condition);
实例2:在查询告警窗口中对告警流水号为2012和2013的记录做“清除”操作
【思路】:
与实例1不同,这里的CSN有两个,意味着告警流水号CSN必须是一个数组,那么向服务端请求的JSON对象应该为{“cmdCode”:41, “csn”:[2012, 2013]}
【演示代码】:
1、改造AlarmClearCondition.java定义,使csn变成一个集合csnList
public class AlarmClearCondition
{
private int cmdCode = 0;
private List<Integer> csnList = newArrayList<Integer>();
public void setCmdCode(int cmdCode)
{
thisNaNdCode = cmdCode;
}
public void setCsn(int csn)
{
csnList.add(csn);
}
}
2、把AlarmClearCondition对象转换为Json对象
AlarmClearConditioncondition = new AlarmClearCondition();
condition.setCmdCode(41);
condition.setCsn(2012);
condition.setCsn(2013);
Gson gson = newGson();
String jsonCond= gson.toJson(condition);
实例3:若实例2中对流水号2012和2013的两条记录“清除”成功后,服务端返回的结果为{“errorCode”:0, “errORMsg”:“Clear alarmssuccessful!”, “clearedCsn”:[2012, 2013]},则客户端如何把这个结果解析为Java对象?
【思路】:
1、直接用硬编码进行解析未尝不可,但代码不够优雅。
2、与生成Json对象一样定义一个Java类,剩余的转换工作由Gson协助完成。见演示代码。
【演示代码】:
1、定义AlarmClearResult类
public class AlarmClearResult
{
private int errorCode = 0;
privateString errorMsg = null;
private List<Integer> clearedCsn = new ArrayList<Integer>();
public int getErrorCode()
{
return errorCode;
}
public String getErrorMsg()
{
return errorMsg;
}
public List<Integer> getClearedCsn()
{
return clearedCsn;
}
}
2、把Json对象转换为AlarmClearResult对象
AlarmClearResult result =gson.fromJson(jsonObject, AlarmClearResult.class);
实例4:当前屏最后一条告警记录流水号为44531611,查询下一屏告警记录时与服务端约定请求条件为{“FieldName”:”logId”,”Operator”:”<”, “value”:44531611},使用Gson把Java对象转换为此Json对象,但要求Java的属性字段定义为是“驼峰”风格。
【思路】:
1、不考虑属性字段定义“驼峰”风格问题,那只需要把FileName、Operator和value作为Java类的属性即可,见演示代码1。
2、为使Java属性字段定义和显示为不同的值,Gson提供了@SerializedName注释,见演示代码2.
【演示代码1】:
1、定义EndlessQuery类
public class EndlessQuery
{
private String FieldName = null;
private String Operator = null;
private int value = 0;
public void setFieldName(String filedName)
{
this.FieldName = fieldName;
}
public void setOperator(String operator)
{
this.Operator = operator;
}
public void setValue(int value)
{
this.value = value;
}
}
2、把AlarmClearResult转换为JSON对象
EndlessQueryqueryCond = new EndlessQuery();
queryCond.setFiledName(“logId”);
queryCond.setOperator(“<”);
queryCond.setValue(44531611);
Gson gson = newGson();
System.out.println(gson.toJson(queryCond));
【演示代码2】:
只修改EndlessQuery类的FileName和Operator两个字段即可,其余保持不变,如下:
public class EndlessQuery
{
@SerializedName(“FieldName”)
private String fieldName= null;
@SerializedName(“Operator”)
private String operator= null;
private int value = 0;
public void setFieldName(String filedName)
{
this.fieldName = fieldName;
}
public void setOperator(String operator)
{
this.operator = operator;
}
public void setValue(int value)
{
this.value = value;
}
}
实例5:列举一个稍微真实的场景,告警监控TOPN是指客户端向任意一个应用服务下发请求后,应用服务从各个数据服务中读取最新上报的TOPN条告警记录,然后再把获取的告警记录按条件进行重排序取TOPN条给客户端。
假如监控TOPN的命令码为22,每次监控40条最新的告警记录,在获取这40条最新的告警记录时使用的排序条件为“按网元发生时间”和“告警流水号”进行降序排序,其JSON对象为{"CmdCode":22, "Limit":40, "VersionID":0, "SortCond":[{"FieldName":"occurNeTime", "SortOrder":"desc","Priority":1},{"FieldName":"logId", "SortOrder":"desc", "Priority":6}]},则客户端如何生成这个JSON对象?
【思路】:
1、直接拼凑字符串代码(不是本文的意图)。
2、定义属性字段为CmdCode、Limit、VersionID和SortCond的类,但SortCond与其它字段不同之处在于SortCond是一个集合,这个集合中对象类型必须能够描述出按“网元发生时间”和“流水号”进行排序。
【演示代码】:
1、定义排序SortConditionBean对象
public class SortConditionBean
{
@SerializedName("FieldName")
private String fieldName = null;
@SerializedName("SortOrder")
private String order = null;
@SerializedName("Priority")
private int priority = 0;
public void setFieldName(String fieldName)
{
this.fieldName = fieldName;
}
public void setOrder(String order)
{
this.order = order;
}
public void setPriority(int priority)
{
this.priority = priority;
}
}
2、定义监控TOPN对象MonitorTopnCond类
public class MonitorTopnCond
{
@SerializedName("CmdCode")
private int cmdCode = 0;
@SerializedName("VersionID")
private int versionID = 0;
@SerializedName("Limit")
private int limit = 0;
@SerializedName("SortCond")
private List<SortCondtionBean> sortCond = new ArrayList<SortConditionBean>():
public void setCmdCode(int cmdCode)
{
this.cmdCode = cmdCode;
}
public void setVersionID(int versionID)
{
this.versionID = versionID;
}
public void setLimit(int limit)
{
this.limit = limit;
}
public void setSortCond(SortConditionBean sortItem)
{
this.sortCond.add(sortItem);
}
}
3、测试方法:
SortConditionBean occuTime = new SortConditionBean();
occuTime.setFieldName("occurNeTime");
occuTime.setOrder("desc");
occuTime.setPriority(1);
SortConditionBean logID = new SortConditionBean();
logID.setFieldName("logId");
logID.setOrder("desc");
logID.setPriority(6);
MonitorTopnCond topnCond = new MonitorTopnCond();
topnCond.setCmdCode(22);
topnCond.setLimit(40);
topnCond.setVersionID(0);
topnCond.setSortCond(occuTime);
topnCond.setSortCond(logID);
Gson gson = new Gson();
System.out.println(gson.toJson(topnCond));
实例6:实例5中向应用服务下发监控TopN请求之后,应用服务从各个数据分区读取数据后,再重排序取了Top40条告警记录返回客户端,结果如下(为简单起见这里把40条告警记录改成1条):
{
"ErrorMsg": "0",
"LastVersionId": 0,
"ResultCode":0,
"TotalNum":24006,
"records":
[
{
"Basic":["链路断连","BTS","NE=10009","NE=10009","Location=Test1, m_strPara=1","",-1,19],
"Paras":[0,0,0,0,0,19,0,0,0,0],
"intExt":[[0,0,0,0,0,0,0,0,0,0],[134, 13528,0,0,4,199,199]],
"strExt":[["","",""], ["","","","","127.0.0,1","",""]]
}
]
}
客户端如何解析这个Json对象呢?
【思路】:
仔细观察返回结果可发现,它由ErrorMsg、LastVersionId、ResultCode、TotalNum、records构成,所以可以定义一个Java类使其包含这五个属性字段。重要的一点records不同于ErrorMsg等属性字段,它又由有Basic、Paras、intExt和strExt构成。
【演示代码】:
1、定义告警记录对象AlarmRecord,对应record集合中的一个元素。
public class AlarmRecord
{
@SerializedName("Basic")
private List<Object> basicField = null;
@SerializedName("Paras")
private List<Integer> parasField = null;
@SerializedName("intExt")
private int[][] intExtField = null;
@SerializedName("strExt")
private String[][] strExtField = null;
public List<Object> getBasicField()
{
return basicField;
}
public List<Integer> getParasField()
{
return parasField;
}
public int[][] getIntExtField()
{
return intExtField;
}
public String[][] getStrExtField()
{
return strExtField;
}
}
2、定义结果响应对象AlarmResponseRecord
public class AlarmResponseRecord
{
@SerializedName("ResultCode")
private int errorCode = 0;
@SerializedName("ErrorMsg")
private String errorMessage = null;
@SerializedName("LastVersionId")
private int versionID = 0;
@SerializedName("TotalNum")
private int totalNum = 0;
private List<ALarmRecord> records = null;
public int getErrorCode()
{
return errorCode;
}
public String getErrorMessage()
{
return errorMessage;
}
public int getVersionID()
{
return versionID;
}
public int getTotalNum()
{
return totalNum;
}
public List<AlarmRecord> getRecords()
{
return records;
}
}
3、测试方法:
String result = "{'ErrorMsg':'ccq', 'LastVersionId':0, 'ResultCode':0, 'TotalNum':24006, 'records':[{'Basic':[0,'通讯系统',3,0,0,0,'链路断连'], 'Paras':[0,0,0,19,0,0], 'intExt':[[1,2,3,4,5],[6,7,8,9,11]], 'strExt':[['a','b','c','d'],['','','']]}]}";
Gson gson = new Gson();
AlarmResponseRecord almRecords = gson.fromJson(result, AlarmResponseRecord .class);
System.out.println(almRecords.getRecords());
实例7:告警监控TOPN请求的JSON对象为:{"CmdCode":22,"Limit":40,"VersionID":0,"SortCond":[{"FieldName":"occurNeTime", "SortOrder":"desc", "Priority":1},{"FieldName":"logId", "SortOrder":"desc","Priority":6}]}
告警无尽列表查询请求的JSON对象为{"CmdCode":21,"Limit":41,"SortCond":[{"FieldName":"occurNeTime", "SortOrder":"desc", "Priority":1},{"FieldName":"logId", "SortOrder":"desc","Priority":6}],"EndlessTableCond":[{"FieldName":"occurNeTime", "Operator":"=","value":[1355899366]},{"FieldName":"logId", "Operator":"<","value":[33394632]}]}
仔细观察会发现这两个请求的JSON对象结构非常类似,如何使用一个Java对象生成这两个Json对象呢?
【思路】:
很容易想到的是把相同的数据作为Java的属性字段(如CmdCode、Limit、SortCond),不同的数据作为子类属性字段(如EndlessTableCond、VersionID),使用Java的继承机制可实现。
【演示代码】:
1、定义排序对象SortConditionBean
public class SortConditionBean
{
@SerializedName("FieldName")
private String fieldName = null;
@SerializedName("SortOrder")
private String order = null;
@SerializedName("Priority")
private int priority = 0;
public void setFieldName(String fieldName)
{
this.fieldName = fieldName;
}
public void setOrder(String order)
{
this.order = order;
}
public void setPriority(int priority)
{
this.priority = priority;
}
}
2、定义无尽列表查询对象EndlessConditionBean
public class EndlessConditionBean
{
private String FieldName = null;
private String Operator = null;
private int value = 0;
public void setFieldName(String fieldName)
{
this.FieldName = fieldName;
}
public void setOperator(String operator)
{
this.Operator = operator;
}
public void setValue(int value)
{
this.value = value;
}
}
3、定义公共父类AlarmConditionBean对象
public class AlarmConditionBean
{
@SerializedName("CmdCode")
private int cmdCode = 0;
@SerializedName("Limit")
private int limit = 0;
@SerializedName("SortCond")
private List<SortConditionBean> sortCond = new ArrayList<SortConditionBean>();
public void setCmdCode(int cmdCode)
{
thisNaNdCode = cmdCode;
}
public void setLimit(int limit)
{
this.limit = limit;
}
public void setSortCond(SortConditionBean sortItem)
{
this.sortCond.add(sortItem);
}
}
5、定义监控AlarmMonitorBean类
public class AlarmMonitorBean extends AlarmConditionBean
{
@SerializedName("VersionID")
private int versionID = 0;
public void setVersionID(int versionID)
{
this.versionID = versionID;
}
}
6、定义查询AlarMQueryBean类
public class AlarmQueryBean extends AlarmConditionBean
{
@SerializedName("EndlessTableCond")
private List<EndlessConditionBean> queryCond = new ArrayList<EndlessConditionBean>();
public void setEndlessCond(EndlessConditionBean endlessItem)
{
queryCond.add(endlessItem);
}
}
7、测试代码
测试告警监控TOPN
SortConditionBean occuTimeSortItem = new SortConditionBean();
occuTimeSortItem.setFieldName("occurNeTime");
occuTimeSortItem.setOrder("desc");
occuTimeSortItem.setPriority(1);
SortConditionBean logIDSortItem = new SortConditionBean();
logIDSortItem.setFieldName("logId");
logIDSortItem.setOrder("desc");
logIDSortItem.setPriority(6);
AlarmMonitorBean monitorCond = new AlarmMonitorBean();
monitorCond.setCmdCode(22);
monitorCond.setLimit(40);
monitorCond.setVersionID(0);
monitorCond.sortSortCond(occuTimeSortItem);
monitorCond.sortSortCond(logIDSortItem);
Gson gson = new Gson();
System.out.println(gson.toJson(monitorCond));
测试告警无尽列表查询:
SortConditionBean occuTimeSortItem = new SortConditionBean();
occuTimeSortItem.setFieldName("occurNeTime");
occuTimeSortItem.setOrder("desc");
occuTimeSortItem.setPriority(1);
SortConditionBean logIDSortItem = new SortConditionBean();
logIDSortItem.setFieldName("logId");
logIDSortItem.setOrder("desc");
logIDSortItem.setPriority(6);
EndlessConditionBean occuTimeEndlessItem = new EndlessConditionBean();
occuTimeEndlessItem.setFieldName("occurNeTime");
occuTimeEndlessItem.setOperator("=");
occuTimeEndlessItem.setValue(1355899366);
EndlessConditionBean logIDEndlessItem = new EndlessConditionBean();
logIDEndlessItem.setFieldName("logId");
logIDEndlessItem.setOperator("<");
logIDEndlessItem.setValue(33394632);
AlarmQueryBean queryCond = new AlarmQueryBean();
queryCond.setCmdCode(21);
queryCond.setLimit(41);
queryCond.setSortCond(occuTimeSortItem);
queryCond.setSortCond(logIDSortItem);
queryCond.setEndlessCond(occuTimeEndlessItem);
queryCond.setEndlessCond(logIDEndlessItem);
Gson gson = new Gson();
System.out.println(gson.toJson(queryCond));
【备注】:
关于此例还有一种解决方式,即使用Gson的@Since注释,但这样会影响代码的易读性,不建议使用,有兴趣的读者可以了解一下Gson的api
实例8:服务端返回给客户端的数据只是一个数组,如[96301,0,1,"链路断连","BTS","NE=1009"],此时客户端如何解析?
【思路】:
JSON对象在定义数据时明确指出数据是一个键值对,但并不表明Gson不能解析,因为这也是一个Json对象,这样做的目的主要是为了减少通讯传输量给带宽带来的拥挤。假如各字段的含义如下:
96301 | 0 | 1 | 链路断连 | BTS | NE=1009 |
告警流水号 | 清除状态 | 确认状态 | 告警类型 | 告警源类型 | 告警源 |
【演示代码】:
1、定义告警记录AlarmRecord类
final class AlarmSimpleRecord
{
private int logID = 0;
private int clearStatus = 0;
private int confirmStatus = 0;
private String alarmType = null;
private String neType = null;
private String neFDN = null;
public int getLogID()
{
return logID;
}
public void setLogID(int logID)
{
this.logID = logID;
}
public int getClearStatus()
{
return clearStatus;
}
public void setClearStatus(int clearStatus)
{
this.clearStatus = clearStatus;
}
public int getConfirmStatus()
{
return confirmStatus;
}
public void setConfirmStatus(int confirmStatus)
{
this.confirmStatus = confirmStatus;
}
public String getAlarmType()
{
return alarmType;
}
public void setAlarmType(String alarmType)
{
this.alarmType = alarmType;
}
public String getNeType()
{
return neType;
}
public void setNeType(String neType)
{
this.netype = neType;
}
public String getNeFDN()
{
return neFDN;
}
public void setNeFDN(String neFDN)
{
this.neFDN = neFDN;
}
}
2、解析演示
public class AlarmRecord
{
private static final int LOG_ID = 0;
private static final int CLEAR_STATUS = 1;
private static final int CONFIRM_STATUS = 2;
private static final int ALARM_TYPE = 3;
private static final int NE_TYPE = 4;
private static final int NE_FDN = 5;
public static void main(String[] args)
{
String result = "[96031, 0 ,1, '链路断连','BTS', 'NE=10009']";
Gson gson = new Gson();
JsonElement resultElement = gson.fromJson(result, JsonElement.class);
if(resultElement.isJsonNull())
{
// do nothing.
return;
}
AlarmSimpleRecord record = new AlarmSimpleRecord();
JsonArray elementArray = resultElement.getAsJsonArray();
record.setLogID(elementArray.get(LOG_ID).getAsInt());
record.setClearStatus(elementArray.get(CLEAR_STATUS).getAsInt());
record.setConfirmStatus(elementArray.get(CONFRIM_STATUS).getAsInt());
record.setAlarmType(elementArray.get(ALARM_TYPE).getAsString());
record.setNeType(elementArray.get(NE_TYPE).getAsString());
record.setNeFDN(elementArray.get(NE_FDN).getAsString());
System.out.println(record);
}
}
相关文章