如何使用GSON和反序列化JsonWriter对象的数组

Question

我有以下类别：

public class MeetingCenter {
    private String name;
    private List<MeetingRoom> meetingRoomList;
}

public class MeetingRoom {
    private MeetingCenter meetingCenter;
    private String code;
    private String name;
    private List<Reservation> reservationList;
}

public class Reservation {
    private MeetingRoom meetingRoom;
    private String owner;
}

我想创建以下架构JSON：

该方法将抛出我的，我叫的toJSON（）方法，行异常：

private static void exportToJson(List<MeetingCenter> mcs) throws IOException {
    Gson gson = new Gson();
    String data = gson.toJson(mcs);

    JsonWriter writer = new JsonWriter(new FileWriter("export.json"));
    writer.setIndent("    "); // set indent

    writer.beginObject(); // document start
    writer.name("schema").value("PLUS4U.EBC.MCS.MeetingRoom_Schedule_1.0");
    writer.name("uri").value("ues:UCL-BT:UCL.INF/DEMO_REZERVACE:EBC.MCS.DEMO/MR001/SCHEDULE");
    writer.name("data").value(data);

    writer.endObject(); // document end
    writer.close();
}

例外：

Exception in thread "main" java.lang.StackOverflowError
    at java.lang.StringBuffer.append(StringBuffer.java:380)
    at java.io.StringWriter.write(StringWriter.java:77)
    at com.google.gson.stream.JsonWriter.beforeName(JsonWriter.java:614)
    at com.google.gson.stream.JsonWriter.writeDeferredName(JsonWriter.java:401)
    at com.google.gson.stream.JsonWriter.beginArray(JsonWriter.java:287)
    at com.google.gson.internal.bind.CollectionTypeAdapterFactory$Adapter.write(CollectionTypeAdapterFactory.java:95)
    ....

Answer 1

你的对象都有他们的父母引用。

GSON在MeetingCenter需要容貌然后尝试序列及其子MeetingRooms。该MeetingRooms有参考回到MeetingCenter，所以GSON善有善报和转转，直到你得到一个堆栈溢出。

为了解决这个问题，你可以确保只露出孩子，而不是父母。有很多展示本已问题。见Java Gson Exclude fields during serialization。

例如，您可能Reservation是这样的：

class Reservation {
    MeetingRoom meetingRoom;
    @Expose
    String owner;
}

我会离开，其余给你。

---

此外，你有一个方法调用，当你还没有开始的数组writer.endArray()。删除了这一行。

writer.beginObject(); // document start
writer.name("schema").value("PLUS4U.EBC.MCS.MeetingRoom_Schedule_1.0");
writer.name("uri").value("ues:UCL-BT:UCL.INF/DEMO_REZERVACE:EBC.MCS.DEMO/MR001/SCHEDULE");
writer.name("data").value(data);

//writer.endArray(); removed
writer.endObject(); // document end
writer.close();

Answer 2

该@expose是避免计算器异常，但语句中的解决方案

writer.name("data").value(data);

是无效的，因为该数据将逃逸字符富集。例如，你可以在数据字段

"data": "{\"name\": \"center 1\" ... }"

所以有可能会在反序列化阶段的问题。

我的实现提出了在模式和URI的可配置MeetingCenter类容器类。

/** Container class configures the schema and URI */
public class Container {
    @Expose
    private String schema;
    @Expose
    private String uri;
    @Expose
    private List<MeetingCenter> data;
}

public class Reservation {
    private MeetingRoom meetingRoom;
    @Expose
    private String owner;
}

public class MeetingRoom {
    private MeetingCenter meetingCenter;
    @Expose
    private String code;
    @Expose
    private String name;
    @Expose
    private List<Reservation> reservationList;
}

public class MeetingCenter {
    @Expose
    private String name;
    @Expose
    private List<MeetingRoom> meetingRoomList;
}

public class Main {
    public static void main(String[] args){
        Container container = meetingCenterInitialization();

        GsonBuilder builder = new GsonBuilder();
        builder.setPrettyPrinting();
        // it is necessary to avoid stackoverflow
        builder.excludeFieldsWithoutExposeAnnotation();

        Gson gson = builder.create();

        String jsonString = gson.toJson(container);
        System.out.println(jsonString);


        Container container1 = gson.fromJson(jsonString, Container.class);
        System.out.println("\n\n\n\n" + container1.getData().get(0).getName());
    }
}

的主要方法的输出是

{
  "schema": "PLUS4U.EBC.MCS.MeetingRoom_Schedule_1.0",
  "uri": "ues:UCL-BT:UCL.INF/DEMO_REZERVACE:EBC.MCS.DEMO/MR001/SCHEDULE",
  "data": [
    {
      "name": "center name",
      "meetingRoomList": [
        {
          "code": "room 1",

         ...