JPA Circular dependency on same entity resulting in recursive output - json

Question

I am having a Folder class which in-turn can have many sub-folders. In this case i am trying to make use of same entity that single entity is being used to achieve the same.
Something like below.
Folders
----SubFolder
----SubFolder
Below are the classes.
FolderData.java
#Entity
#JsonIdentityInfo(generator =ObjectIdGenerators.IntSequenceGenerator.class,property="projectId")
public class FolderData {
#Id
#SequenceGenerator(name = "seq-gen", initialValue = 1)
#GeneratedValue(strategy = GenerationType.IDENTITY, generator = "seq-gen")
private Integer parentId;
private int moduleId;
private int subProjectId;
private String folderName;
private Integer folderId;
private int projectId;
#ManyToOne(fetch = FetchType.LAZY,cascade=CascadeType.ALL)
#JoinColumn(name="folder_child")
#JsonIgnore
private FolderData folderData;
#OneToMany(mappedBy = "folderData")
#JsonIgnoreProperties("folderList")
private Set<FolderData> folderList=new HashSet<>();
}
FodlerController.java
#RestController
#RequestMapping("/folder")
public class FodlerController {
#Autowired
private FolderService folderService;
#GetMapping(produces = "application/json")
public List<FolderData> getFolderList(){
return folderService.findAllFromTheList();
}
#PostMapping(produces = "application/json", consumes = "application/json")
public void createFolder(#RequestBody FolderData folderData) {
if(folderData.getId()==null && folderData.getFolderId()==null) {
System.out.println("id is null");
folderData.setFolderId(new Random().nextInt());
folderService.save(folderData);
}
else {
folderService.doChildAddition(folderData);
}
}
}
FolderService.java
#Service
public class FolderService {
#Autowired
private FolderRepo folderRepo;
public FolderData save(FolderData folderData) {
return folderRepo.save(folderData);
}
public FolderData getFolderDataByParentId(Integer id) {
return folderRepo.getOne(id);
}
public List<FolderData> findAllFromTheList() {
return folderRepo.findAll();
}
public FolderData getFolderDataByfolderId(Integer folderId) {
return folderRepo.findFolderByFolderId(folderId);
}
public void doChildAddition(FolderData childFolder) {
FolderData parentFolder=folderRepo.findFolderByFolderId(childFolder.getFolderId());
childFolder.setFolderData(parentFolder);
FolderData childFolderSaved = folderRepo.saveAndFlush(childFolder);
//parentFolder.getFolderList().add(folderRepo.getOne(childFolderSaved.getId()));
//folderRepo.save(parentFolder);
}
}
Json Request for creating parent and the response collected which includes autogenerated folder id:
{
"moduleId":1,
"subProjectId":1,
"folderName":"One",
"projectId":1
}
[
{
"projectId": 0,
"moduleId": 1,
"subProjectId": 1,
"folderName": "One",
"folderId": 963031296,
"folderList": [],
"id": 1
}
]
Creating child for the parent and the response collected.
{
"moduleId":1,
"subProjectId":2,
"folderName":"Two",
"projectId":1,
"folderId": -963031296
}
[
{
"projectId": 0,
"moduleId": 1,
"subProjectId": 1,
"folderName": "One",
"folderId": 963031296,
"folderList": [
{
"projectId": 0,
"moduleId": 1,
"subProjectId": 2,
"folderName": "Two",
"folderId": 963031296,
"id": 2
}
],
"id": 1
},
2
]
For the above response i am getting the No 2 with the response since i am using
#JsonIdentityInfo(generator =ObjectIdGenerators.IntSequenceGenerator.class,property="parentId")
else the whole object would have come instead of 2.
Example output had i not used JsonIdentityInfo
[
{
"projectId": 0,
"moduleId": 1,
"subProjectId": 1,
"folderName": "One",
"folderId": 963031296,
"folderList": [
{
"projectId": 0,
"moduleId": 1,
"subProjectId": 2,
"folderName": "Two",
"folderId": 963031296,
"id": 2
}
],
"id": 1
},
{
"projectId": 0,
"moduleId": 1,
"subProjectId": 2,
"folderName": "Two",
"folderId": 963031296,
"id": 2
}
]

Answer 1

It's easy, #Prash
Since you have used a Set in a undirectional dependency you must supply an uniqueness of a joined (depended) entities. Any set is using hashCode() and equals() to determine an availability of the concurent item that is want to be added. Your FolderData class doesn't have hashCode() and equals() so from the point of view of JVM all the deserialized objects are same and because of the Set nature it's normal to keep only last example