i have an entity with a unique name.
In my example i save two persons with the same name. At the second time comes an "EntityExists" (Unique) Exception, that was the expected behavior.
After it i changed name and set the "ID" to null.
Than i try to persist it again but i get "org.apache.openjpa.persistence.EntityExistsException: Attempt to persist detached object "com.Person#1117a20". If this is a new instance, make sure any version and/or auto-generated primary key fields are null/default when persisting.
without the version it works but i find no solution to "reset" the version number.
Can someone help me?
Update: My new problem is, that i have a base entity an two pcVersionInit (look at my answer at bottom) i can't override it, i tried it in base and normal entity what is the best practise now instead of "override" the value in pcVersionInit ? Copy Constructor?"
public class Starter{
private static EntityManager em;
public static void main(String[] args) {
em = Persistence.createEntityManagerFactory("openjpa")
.createEntityManager();
Person p1 = new Person("TEST");
savePerson(p1);
Person p2 = null;
try{
p2 = new Person("TEST");
savePerson(p2);
}catch(Exception e){
p2.setId(null);
p2.setName(p2.getName()+"2");
em.persist(p2);
}
}
private static void savePerson(Person person){
em.getTransaction().begin();
em.persist(person);
em.getTransaction().commit();
}
}
Person.class:
#Entity
public class Person implements Serializable{
#Id
#GeneratedValue(strategy = GenerationType.SEQUENCE,generator="seqGenerator")
#SequenceGenerator(name="seqGenerator",sequenceName="personSeq")
private Long id;
#Version
private Long version;
#Column(nullable=true, unique=true)
private String name;
public Person(String name) {
this.name = name;
}
public Long getId() {
return id;
}
public void setId(Long id) {
this.id = id;
}
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
public Long getVersion() {
return version;
}
public void setVersion(Long version) {
this.version = version;
}
}
First off, stay away from messing around with pcVersionInit. I'd suggest to create a copy constructor in your Person Entity and in the event of rollback create a new one using the copy constructor.
Okay the problem is that OpenJPA adds a field named pcVersionInit (with #version) and set it "true" after try to persist. If i use reflection to set it to false, it works. Other way is a copy constructor.
Related
Let's assume I have many teams like : "Team A, Team B...Team Z" and each team has at least 5 components. Now I want to create a generic controller that responds to any request of type /team/number so I can be able to get informations about a team member.
For example my controller must be able to map this request :
#RequestMapping(value = "/Team A/1", method = RequestMethod.GET)
Team class could be :
#Entity
#Table(name = "team")
public class Team {
public Team() {}
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
#Id
#GeneratedValue(strategy = GenerationType.AUTO)
private Long id;
private String name;
private List<Player> players;
}
And
Player :
#Entity
#Table(name = "player")
public class Player {
public Player() {}
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
#Id
#GeneratedValue(strategy = GenerationType.AUTO)
private Long id;
private String name;
private int number;
}
Obviously it can either execute GET and POST. The point is that I don't want to specify a controller for each team and each number, I just want one that can respond to /String/int .
I need also to specify which string it can accept and the range of values (max 5 for example).
The #PathVariable spring annotation would help you in this case.
For example:
#RequestMapping(value = "{team}/{component}", method = RequestMethod.GET)
public Team getTeam(#PathVariable String team, #PathVariable int component) {
//Use team and component here
}
You can just add some logic into the controller method to throw some exception if either variable is an unexpected value.
Though depending on what you are trying to achieve, this may be a bad design as this controller method will catch all GET requests that match String/int. You may want to try something like this to make the request mapping more specific:
#RequestMapping(value = "team/{teamLetter}/{component}", method = RequestMethod.GET)
public Team getTeam(#PathVariable char teamLetter, #PathVariable int component) {
//Use team and component here
}
I have just started experimenting with Neo4J/Java and expect this is an easy one I'm missing, and probably phrasing my queries wrong.
I have some model classes as follows:
#Node
public class Garment {
#Id
#GeneratedValue
private Long id;
private String name;
#Relationship(type = "DESIGNED_BY")
private Entity designer;
// Other properties getters/setters removed for readibility
public Entity getDesigner() {
return designer;
}
public void setDesigner(Entity designer) {
this.designer = designer;
}
}
public abstract class Entity {
#Id
#GeneratedValue
Long id;
private String name;
public Long getId() {
return id;
}
public void setId(Long id) {
this.id = id;
}
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
}
#Node
public class Person extends Entity {
}
#Node
public class Company extends Entity {
}
And a corresponding repository
public interface Garment extends Neo4jRepository<Garment, Long> {
Garment findByName(String name);
}
I have no problem inserting, using repository.save(), this correctly adds everything; nodes, relationships. Fine. It gives Designers of type Person labels of Person, and Designers of type Company the label Company.
However, when I do a find, e.g. findByName(), findAll(). it is not matching the designer and just saying designer is null, according to the cipher being executed/logged it looks like it's trying to build a relationship there with nodes with an Entity label, which there are none.
How can I get my repository to return Garments with designers of Person and Companys. I expect this is going to be as simple as an annotation, in order to fix.
(Note I've tried adding a #Node on the entity type with Person and Company as labels, however it just results in every node being added as both a Person and a Company).
The question here is why the entity properties are not being saved after calling some setters of the entity. Usually when changing a property of a managed entity, it should propagate to the database.
Take a look at this example:
#Service
public class SystemServiceImpl implements SystemService {
#Autowired
private SystemDao systemDao;
#Override
#Transactional
public System replace(Long systemID) {
// External system to replace
System system = systemDao.findByID(systemID);
if (null != system) {
system.setName("Test"); // Calling findByID again shows that this call did not have any effect.
}
return system;
}
}
-
#Entity
#Table(name = "db.system")
public class System {
#Id
#GeneratedValue(strategy = GenerationType.IDENTITY)
private Long systemID;
private String name;
#OneToMany(mappedBy = "system", fetch = FetchType.LAZY)
#JsonIgnore
private List<Customer> customers = new ArrayList<Customer>();
public Long getSystemID() {
return systemID;
}
public void setSystemID(Long systemID) {
this.systemID = systemID;
}
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
public List<Customer> getCustomers() {
return customers;
}
public void setCustomers(List<Customer> customers) {
this.customers = customers;
}
}
If I call systemDao.merge after the system.setName("Test") then it is saved to the database. I feel I should not have to call merge as this should be a managed entity.
I tried having the replace method both with #Transactional and without, and both are producing the same result.
Any ideas?
Thanks,
The behaviour can be explained if SystemDao.findByID() returns a detached object. Make sure that SystemDao does not use RequiresNew-Transaction or explicitly detach the object after loading it.
It would really help if you posted the code of SystemDao and any relevant configuration entries (Spring, entity manager config, ...).
The error was happening when I was running my integration tests. Turns out that Transactional is not doing anything because there is no session. The solution to this was to add a Transactional to my tests.
I've been using Spring Data for saving entities to the mongo DB and my code at the moment looks like this:
I have a repo class:
public interface LogRepo extends MongoRepository<Log, String> {
}
and I have an Entity Log which looks like this:
#Document(
collection = "logs"
)
public class Log {
#Id
private String id;
private String jsonMessage;
public String getId() {
return id;
}
public void setId(String id) {
this.id = id;
}
public String getJsonMessage() {
return jsonMessage;
}
public void setJsonMessage(String jsonMessage) {
this.jsonMessage = jsonMessage;
}
}
and this work well for me, however this approach works only for the case if I want to save Log entities to "logs" collection. However it would be very nice for me to be able to save Log entity to different collections depending on the context. I mean it would be nice to define collection name in the runtime. Is it possible somehow?
Thanks, cheers
Try to use inheritance and define appropriate collection names in such way. May give you possibility to save in different collections but you will be still not able to specify dynamically collection names and resp. their amount at runtime.
#Document(
collection = "logs"
)
public class Log {
#Id
private String id;
private String jsonMessage;
public String getId() {
return id;
}
public void setId(String id) {
this.id = id;
}
public String getJsonMessage() {
return jsonMessage;
}
public void setJsonMessage(String jsonMessage) {
this.jsonMessage = jsonMessage;
}
}
#Document(
collection = "log_child"
)
public class LogChild extends Log{}
With the MongoOperations save method you can choose which class to use and
based on the class it will choose the appropriate collection.
#Document(collection = "collection_#{T(com.github.your_project.DBUtils).getCollectionName()}")
public Class Collection
You can change the name in real time using a static getter
#UtilityClass
public class DBUtils {
private String collectionName;
public String getCollectionName() {
return collectionName;
}
public void setCollectionName(String collectionName) {
DBUtils.collectionName = collectionName;
}
}
Well i want to know if there is a much appropriate way to tackle generating auto id with string values, my first idea is creating an auto increment id which we can call auto_id then before saving a new entity I'll query for the latest data inside the db to get the id then I'll add 1 to my auto generate value column that I assign name which is stringValue+(id+1) though I'm concerned on how it will affect the performance as to saving this entity needs two access in db which is fetching and saving... like my question earlier is there a much appropriate way to handle this scenario?
And also sorry for my English guys if you want to clarify things with my question kindly ask, thnx in advance..
Here's my code for AttributeModel for hibernate annotation
#Component
#Entity
#Table(name="attribute_info")
public class AttributeModel {
#Id
#GeneratedValue(strategy=GenerationType.AUTO)
#Column(name="attr_id", nullable=false, unique=true)
private int id;
#Column(name="attr_name")
private String name;
#Column(name="attr_desc")
private String desc;
#Column(name="attr_active")
private int active;
#Column(name="attr_abbr")
private String abbr;
#OneToOne(cascade = CascadeType.ALL, fetch = FetchType.EAGER)
#JoinColumn(name="stats_id", referencedColumnName="stats_id")
private BaseStatisticModel baseStats;
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
public String getDesc() {
return desc;
}
public void setDesc(String desc) {
this.desc = desc;
}
public int getActive() {
return active;
}
public void setActive(int active) {
this.active = active;
}
public String getAbbr() {
return abbr;
}
public void setAbbr(String abbr) {
this.abbr = abbr;
}
public BaseStatisticModel getBaseStats() {
return baseStats;
}
public void setBaseStats(BaseStatisticModel baseStats) {
this.baseStats = baseStats;
}
public int getId() {
return id;
}
public void setId(int id) {
this.id = id;
}
}
I can only say "Don't do it". How is a String ID like "str10001" better than 10001? It can't be an optimization as strings take more memory and more time. So I guess you need to pass it to some String-expecting method later.
If so, then pass "str" + id instead. Constructing the string on the fly surely won't saturate your server.
If not, then let us know what you actually need rather than what you think it could help you to achieve it.
I'm pretty sure, Hibernate can't do it. It couldn't some long time ago I checked it recently and it makes no sense (in any case, it's not a feature crowds would request).