I have manyToMany relationship between Newsfeed and User in order to store favorite Newsfeeds of Uesrs.
Newsfeed
id
title
description
1
Test Title
Here is description
2
Title 2
Here is another description
Favorite_Newsfeed (ManyToMany Relationship)
newsfeed_id
user_id
1
1
2
2
1
2
Now I need to retrieve all the newsfeeds along with favorites filtered by user.
NewsfeedWithFavorite (Goal)
id
title
description
isFavorite
1
Test Title
Here is description
1
2
Title 2
Here is another description
I can do this by using case when in raw query but What is the best approach to retrieve data in this format using JPA? (I am new in JPA)
You have to use join in JPQL and then map favorites.
Repository :
public interface NewsFeedJpaRepository extends JpaRepository<NewsFeed, Long> {
#Query("SELECT nf , fnf.user.id FROM NewsFeed nf LEFT JOIN FavoriteNewsFeed fnf ON fnf.newsfeed = nf and fnf.user.id = :userId ")
Page<Object[]> findAllWithFavorite(Long userId, Pageable pageable);
}
create a dto like following:
public class NewsFeedWithFavorite extends NewsFeed {
private Boolean isFavorite;
public NewsFeedWithFavorite(){};
public NewsFeedWithFavorite(NewsFeed newsFeed, Long userID){
setId(newsFeed.getId());
setTitle(newsFeed.getTitle());
setDescription(newsFeed.getDescription());
if(userID !=null)
setFavorite( Boolean.TRUE);
else
setFavorite(Boolean.FALSE);
};
public Boolean getFavorite() {
return isFavorite;
}
public void setFavorite(Boolean favorite) {
isFavorite = favorite;
}
}
And finally in the Service :
public Page<NewsFeedWithFavorite> findAllWithFavorite(Long userId, Pageable pageable) {
Page<Object[]> result = newsFeedJpaRepository.findAllWithFavorite(userId,pageable);
return result.map(objArr -> new NewsFeedWithFavorite((NewsFeed) objArr[0], (Long) objArr[1]));
}
Do not use #ManyToMany annotation.
It will be extremely hard for you to add multiple columns and in general work with the auto generated Favorite_Newsfeed table.
The best way to handle this scenarior is to use a different table for User's news feed.
Check out the following structure.
public class User {
#OneToMany
private List<UserNewsFeed> newsFeeds;
}
public class NewsFeed {
public String description;
public String title;
}
public class UserNewsFeed {
#ManyToOne
private User user;
#ManyToOne
private NewsFeed newsFeed;
private boolean isFavorite;
}
Related
Taking a reference from link: Postgres to fetch the list having comma separated values, I want to write a query which somhow brings Employee email Table fields as a list for a particular Empployee. This is needed for Spring Batch to Simply match it from the Resultset and create a POJO/Model class like List emails for Employee class?
Can this be possible ?
select c.*, ce.*, string_agg(ce.email, ',') as emails
from root.employee c
full outer join root.employee_email ce
on c.employee_id = ce.employee_id
group by
c.employee_id, ce.employee_email_id
order by
c.employee_id
limit 1000
offset 0;
Your problem is a common one in the batch processing realm and with Spring Batch it is called "Driving Query Based ItemReaders", you can find more about that in here.
Basically you retrieve the Contacts in your reader, and in your processor you add the list of Emails to them.
#Bean(destroyMethod = "")
public JdbcCursorItemReader<Employee> employeeReader(DataSource dataSource) {
JdbcCursorItemReader<Employee> ItemReader = new JdbcCursorItemReader<>();
ItemReader.setDataSource(dataSource);
ItemReader.setSql("SELECT * FROM employee.employee C ");
ItemReader.setRowMapper(new EmployeeRowMapper());
return ItemReader;
}
#Bean
public ItemProcessor<Employee, Employee> settlementHeaderProcessor(JdbcTemplate jdbcTemplate){
return item -> {
root.EMPLOYEE_EMAIL CE WHERE ce.employee_id = ? ",
new Object[]{item.getId()},
new RowMapper<String>() {
#Override
public String mapRow(ResultSet resultSet, int i) throws SQLException {
return resultSet.getString("EMAIL");
}
});
item.setEmails(emails);
return item;
};
}
PS : this could have some performance issues if you have lots of contacts, because for each contact Item you will hit the database to retrieve Emails.
There is another optimized way, by creating a custom reader that will return a List of Contacts (For example 1000 by 1000), and a processor that will enrich them with their emails. This way you will hit the database again for each 1000 Contact Item.
In your reader your retrieve a list of unique Employees page per page (Say your page is 1000 long).
And in your processor for the 1000 employees you retrieve all their emails in one query.
Then for each employee you set the emails retrieved in the last query.
An example might like the following:
public interface EmployeeRepository extends PagingAndSortingRepository<Employee, Integer> {
}
#Getter
class EmployeeVO {
private Long employeeId;
private String email;
EmployeeVO(Long employeeId, String email) {
this.employeeId= employeeId;
this.email = email;
}
}
public class EmployeeListReader implements ItemReader<List<Employee>> {
private final static int PAGE_SIZE = 1000;
private final EmployeeRepository employeeRepository;
private int page = 0;
public EmployeeListReader(EmployeeRepository employeeRepository) {
this.employeeRepository = employeeRepository;
}
public List<Employee> read() throws Exception {
Page<Employee> employees = employeeRepository.findAll(PageRequest.of(page, PAGE_SIZE));
page++;
return employees.getContent();
}
}
#Bean
EmployeeListReader reader(){
return new EmployeeListReader(this.employeeRepository);
}
#Bean
public ItemProcessor<List<Employee>, List<Employee>> settlementHeaderProcessor(NamedParameterJdbcTemplate namedParameterJdbcTemplate) {
return item -> {
List<Long> employeesIds = item.stream().map(Employee::getId).collect(Collectors.toList());
SqlParameterSource parameters = new MapSqlParameterSource("ids", employeesIds);
List<ContactVO> emails = namedParameterJdbcTemplate
.query("SELECT CE.employeeId, CE.EMAIL FROM employee_EMAIL CE WHERE ce.contact_id IN (:ids) ",
parameters,
new RowMapper<ContactVO>() {
#Override
public ContactVO mapRow(ResultSet resultSet, int i) throws SQLException {
return new ContactVO(
resultSet.getLong("EMPLOYEE_ID"),
resultSet.getString("EMAIL"));
}
});
Map<Long, List<ContactVO>> emailsByContactId = emails.stream().collect(Collectors.groupingBy(ContactVO::getContactId));
List<Employee> newEmployeesWithEmails = Collections.unmodifiableList(item);
newEmployeesWithEmails.forEach(employee -> {
employee.setEmails(emailsByContactId.get(employee.getId()).stream().map(ContactVO::getEmail).collect(Collectors.toList()));
});
return newEmployeesWithEmails;
};
}
Hope this helps
I'm new to Mongodb coming from relational databases and I'd also like to point out I'm using SpringBoot with JPA. If I were to build an automotive classified site where I would have thousands of Users and 100s of thousands of listings, how would I go about setting up the schema? I've read some articles that say normalizing nosql data is bad practices.
Anyhow lets say we have the following structure.
User
id
name
email
Cars
id
make
model
year
I would need to be able to list many cars with the User and what i've seen in my examples is it creates a nested array of cars within User. This would work great for user accounts where I'd like to provide the user with all their cars
Where I get a bit confused is with the cars. The cars need to be able to be searched very quickly and would not need the user info right away. In a sql db I would typically do a search against the cars (year, make, model) and grab the user later on if I needed it.
In mongodb, do you create a User document that contains a nested car array? or do you somehow create 2 documents that are both automatically maintained and search against the car document for performance reasons?
Sample code
#Document(collection = "person")
public class Person {
#Id
private String id;
private String firstName;
private String lastName;
// #DBRef(lazy = true)
private List<Listing> listings;
public Person(String firstName, String lastName) {
this.firstName = firstName;
this.lastName = lastName;
}
}
#Document(collection = "listing")
public class Listing {
#Id
public String id;
public String year;
public String make;
public String model;
public String trim;
public Listing(String year, String make, String model, String trim) {
this.year = year;
this.make = make;
this.model = model;
this.trim = trim;
}
}
#Override
public void run(String... args) throws Exception {
repository.deleteAll();
List<Listing> listings = new ArrayList<>();
Listing listing = new Listing("2008", "Ford", "Focus", "SE");
//listingRepository.save(listing);
listings.add(listing);
Person person = new Person("Alice", "Smith");
person.setListings(listings);
// save a couple of customers
repository.save(person);
person = new Person("Bob", "Smith");
listings = new ArrayList<>();
listings.add(new Listing("2018", "Chrysler", "300", "S"));
person.setListings(listings);
repository.save(person);
// fetch all customers
System.out.println("Customers found with findAll():");
System.out.println("-------------------------------");
for (Person _person : repository.findAll()) {
System.out.println(_person);
}
System.out.println();
// fetch an individual customer
System.out.println("Person found with findByFirstName('Alice'):");
System.out.println("--------------------------------");
System.out.println(repository.findByFirstName("Alice"));
System.out.println("Persons found with findByLastName('Smith'):");
System.out.println("--------------------------------");
for (Person _person : repository.findByLastName("Smith")) {
System.out.println(_person);
}
List<Listing> _listings = listingRepository.findAll();
System.out.println("listings " + _listings.size());
_listings.forEach(v -> {
System.out.println(v.toString());
});
}
Going by your entity model,I think what you are looking for is analogous to Many to Many/One to Many relationship in a relational database. So you can go for One way Embedding or Two way Embedding in MongoDb.
For One way embedding, You can create a Car collection like below:
db.carCollection.insertMany([{
_id:1,
make: 'porcha',
model:'qwerty',
year:'2018'
},
{
_id:2,
make: 'ferrara',
model:'uiop',
year:'2018'
}])
You can then go on to create user collection as below:
db.userCollection.insert({
_id:1,
user:'Tom',
email:'tom#tom.com',
car_ids:[1,2]
})
The car_ids is an array which will hold the ids of cars that belong to the user.
You can fetch the cars belonging to an user as(using findOne to fetch the user. Search parameter should be an unique id. I am considering email to be unique here.Ideally it should be user's id) :
var user=db.userCollection.findOne({email:'tom#tom.com'})
db.carCollection.find({_id:{$in:user.car_ids}})
This will fetch you all the cars per user
For fetching cars only you can simply do:
db.carCollection.find({})
For Two way embedding you can have similar array (as in user collection) inside cars collection so that each car can be identified to its user.
Suggest, I have next structure of objects:
class MovieWrapper {
private Movie movie;
}
class Movie {
private User user;
#Enumerated(EnumType.STRING)
private Status status;
}
and I want to query from repository only first occurrence of MovieWrapper by User Id and where status is Active or Pending:
#Repository
public interface MovieWrapperRepository extends JpaRepository<MovieWrapper, Long> {
MovieWrapper findFirstByMovieUserIdAndMovieStatusActiveOrMovieStatusPending(Long userId);
}
How to do it in correct way?
To get 'custom' object from the repo method you should use projections (interface-base projection is preferable IMO).
To implement complex method you can build its query yourself, for example:
public interface MovieRepo extends JpaRepository<Movie, Long> {
Query("select m as movie from Movie m where m.user.id = ?1 and (m.status = 'ACTIVE' or m.status = 'PENDING')")
List<MovieProjection> findActiveOrPending(Long userId, Status status);
}
Where MovieProjection is a simple interface:
public interface MovieProjection {
Movie getMovie();
}
Note to the alias m as movie in the query, it's recommended to use with projections to avoid some errors.
UPDATED
If you need only one records you can use a trick with Pageable:
public interface MovieRepo extends JpaRepository<Movie, Long> {
Query("select m as movie from Movie m where m.user.id = ?1 and (m.status = 'ACTIVE' or m.status = 'PENDING')")
List<MovieProjection> findAllActiveOrPending(Long userId, Status status, Pageable pageable);
default Optional<MovieProjection> getFirstOne(Long userId, Status status) {
return findAllActiveOrPending(userId, status, PageRequest.of(0, 1)).stream().findAny();
}
}
Here we take first page with one record (PageRequest.of(0, 1) - see here)
I have a provider:user=1:N association modeled with entities, Hibernate/JPA.
Then I want to query a provider/user pair via restrictions on attributes of the dependent entity, like certain values for the attributes of the user, let’s say its id, date of birth, etc.
The logged sql has a proper join and all the attributes of the two entities in the select. I tried it manually, it returns the expected single row.
Thus, on entity level, I expect a single provider entity to be returned with the user list containing the queried user.
Indeed, the corresponding provider entity is returned, but when I then want to access the user via the provider’s user list, it hits the DB a second time and reads all users of the provider totally neglecting my restrictions of the query.
The observed behavior is the same for queries formulated with HQL, Hibernate Criteria (also with #Filter), JPA CriteriaBuilder.
What am I missing here?
Do those restrictions only affect the selection of the root entities (which is provider in my case)?
The problem is sketched on
https://docs.jboss.org/hibernate/orm/3.3/reference/en/html/querycriteria.html
Under 5.4 Associations it says:
The kittens collections held by the Cat instances returned by the previous two queries are not pre-filtered by the criteria. If you want to retrieve just the kittens that match the criteria, you must use a ResultTransformer.
Is this thus the intended behavior for this kind of API?
Or is there a convenient possibility to access just the restricted sub set of the dependent entities?
Regards,
Wolfgang
Here comes some source code to precise my verbal description above.
Provider and User table.
drop table EX_USER;
drop table EX_PROVIDER;
create table EX_PROVIDER
( id number(*,0) not null
,name varchar2(255) not null
,primary key (id)
);
insert into EX_PROVIDER (id,name) values (0 ,'Provider_A');
insert into EX_PROVIDER (id,name) values (1 ,'Provider_B');
commit;
create table EX_USER
( id number(*,0) not null
, ex_provider_id number(*,0) not null
,name varchar2(255)
,location varchar2(255)
,primary key (id)
,constraint ex_user_provider_fk foreign key(ex_provider_id) references EX_PROVIDER(id)
);
insert into EX_USER (id,ex_provider_id,name,location) values (0,0,'User_1','Munich');
insert into EX_USER (id,ex_provider_id,name,location) values (1,0,'User_2','Berlin');
insert into EX_USER (id,ex_provider_id,name,location) values (2,1,'User_3','Munich');
commit;
Entities generated with Eclipse "JPA Tools".
#Entity
#Table(name="EX_PROVIDER")
#NamedQuery(name="ExProvider.findAll", query="SELECT e FROM ExProvider e")
public class ExProvider implements Serializable {
private static final long serialVersionUID = 1L;
#Id
#SequenceGenerator(name="EX_PROVIDER_ID_GENERATOR", sequenceName="KONST_SD_SEQ")
#GeneratedValue(strategy=GenerationType.SEQUENCE, generator="EX_PROVIDER_ID_GENERATOR")
private long id;
private String name;
//bi-directional many-to-one association to ExUser
#OneToMany(mappedBy="exProvider",fetch=FetchType.LAZY)
private Set<ExUser> exUsers=new HashSet<ExUser>();
public ExProvider() {
}
public long getId() {
return this.id;
}
public void setId(long id) {
this.id = id;
}
public String getName() {
return this.name;
}
public void setName(String name) {
this.name = name;
}
public Set<ExUser> getExUsers() {
return this.exUsers;
}
public void setExUsers(Set<ExUser> exUsers) {
this.exUsers = exUsers;
}
public ExUser addExUser(ExUser exUser) {
getExUsers().add(exUser);
exUser.setExProvider(this);
return exUser;
}
public ExUser removeExUser(ExUser exUser) {
getExUsers().remove(exUser);
exUser.setExProvider(null);
return exUser;
}
}
#Entity
#Table(name="EX_USER")
#NamedQuery(name="ExUser.findAll", query="SELECT e FROM ExUser e")
public class ExUser implements Serializable {
private static final long serialVersionUID = 1L;
#Id
#SequenceGenerator(name="EX_USER_ID_GENERATOR", sequenceName="KONST_SD_SEQ")
#GeneratedValue(strategy=GenerationType.SEQUENCE, generator="EX_USER_ID_GENERATOR")
private long id;
private String location;
private String name;
//bi-directional many-to-one association to ExProvider
#ManyToOne(fetch=FetchType.LAZY)
#JoinColumn(name="EX_PROVIDER_ID")
private ExProvider exProvider;
public ExUser() {
}
public long getId() {
return this.id;
}
public void setId(long id) {
this.id = id;
}
public String getLocation() {
return this.location;
}
public void setLocation(String location) {
this.location = location;
}
public String getName() {
return this.name;
}
public void setName(String name) {
this.name = name;
}
public ExProvider getExProvider() {
return this.exProvider;
}
public void setExProvider(ExProvider exProvider) {
this.exProvider = exProvider;
}
}
My intention with the below code was to retrieve Provider 0 (Provider_A) containing User 1 (Berlin) in its user list.
#SuppressWarnings("unchecked")
private void demo() {
EntityManager em = null;
try {
em = emf.createEntityManager();
EntityTransaction tx = em.getTransaction();
try {
tx.begin();
Session session = (Session) em.getDelegate();
Criteria crit = session.createCriteria(ExProvider.class, "provider")
.createCriteria("exUsers", "user")
.add(Restrictions.eq("user.location","Berlin"))
;
List<ExProvider> providerList=(List<ExProvider>)crit.list();
logExProviderList(providerList);
tx.commit();
} finally {
if (tx!=null && tx.isActive()) tx.rollback();
}
} finally {
if (em!=null) em.close();
}
}
private void logExProviderList(List<ExProvider> providerList) {
for (ExProvider provider: providerList) {
logger.info("Criteria: provider=["+provider.getId()+"]");
for (ExUser user : provider.getExUsers()) {
logger.info("Criteria: user=["+user.getId()+"] name=["+user.getName()+"] location=["+user.getLocation()+"]");
}
}
}
The 1st SQL is what I expected. It is executed when crit.list() is called. On SQL level it returns the expected single row for Provider 0, User 1.
SELECT this_.id AS id43_1_,
this_.name AS name43_1_,
user1_.id AS id44_0_,
user1_.EX_PROVIDER_ID AS EX4_44_0_,
user1_.location AS location44_0_,
user1_.name AS name44_0_
FROM EX_PROVIDER this_
INNER JOIN EX_USER user1_
ON this_.id=user1_.EX_PROVIDER_ID
WHERE user1_.location=?;
However, this is not mapped to entity level as I expected. There, the restriction seems to affect the selection of the Provider only. When the user list is accessed, all users of the provider are read from DB neglecting the 'Berlin' restriction.
This was the same whether I used HQL, Hibernate Criteria (also with #Filter), JPA CriteriaBuilder.
SELECT exusers0_.EX_PROVIDER_ID AS EX4_43_1_,
exusers0_.id AS id1_,
exusers0_.id AS id44_0_,
exusers0_.EX_PROVIDER_ID AS EX4_44_0_,
exusers0_.location AS location44_0_,
exusers0_.name AS name44_0_
FROM EX_USER exusers0_
WHERE exusers0_.EX_PROVIDER_ID=?;
The log.
Criteria: provider=[0]
Criteria: user=[1] name=[User_2] location=[Berlin]
Criteria: user=[0] name=[User_1] location=[Munich]
I may achieve the expected result set using setResultTransformer(), but in this case the properly selected result is returned as some rows of entities.
#SuppressWarnings("unchecked")
private void demo() {
EntityManager em = null;
try {
em = emf.createEntityManager();
EntityTransaction tx = em.getTransaction();
try {
tx.begin();
Session session = (Session) em.getDelegate();
Criteria crit = session.createCriteria(ExProvider.class, "provider")
.createCriteria("exUsers", "user")
.add(Restrictions.eq("user.location","Berlin"))
.setResultTransformer(Criteria.ALIAS_TO_ENTITY_MAP);
;
List<ExProvider> providerList=(List<ExProvider>)crit.list();
logExProviderMapList(providerList);
tx.commit();
} finally {
if (tx!=null && tx.isActive()) tx.rollback();
}
} finally {
if (em!=null) em.close();
}
}
private void logExProviderMapList(List providerList) {
Iterator iter = providerList.iterator();
while ( iter.hasNext() ) {
Map map = (Map) iter.next();
ExProvider provider = (ExProvider) map.get("provider");
if(provider!=null) {
logger.info("Criteria: provider=["+provider.getId()+"]");
}
ExUser user = (ExUser) map.get("user");
if(user!=null) {
logger.info("Criteria: user=["+user.getId()+"] name=["+user.getName()+"] location=["+user.getLocation()+"]");
}
}
}
The SQL is the same as the 1st SQL above.
SELECT this_.id AS id43_1_,
this_.name AS name43_1_,
user1_.id AS id44_0_,
user1_.EX_PROVIDER_ID AS EX4_44_0_,
user1_.location AS location44_0_,
user1_.name AS name44_0_
FROM EX_PROVIDER this_
INNER JOIN EX_USER user1_
ON this_.id=user1_.EX_PROVIDER_ID
WHERE user1_.location=?;
The log.
Criteria: provider=[0]
Criteria: user=[1] name=[User_2] location=[Berlin]
My question was and is, whether it is possible to get this result as proper entity tree with Provider_A containing exactly the User from Berlin in its list.
Thanks again,
Wolfgang B
I have the following Play Framework entity (using Morphia for persistence) as part of a generic blogging app:
#Entity
public class Comment extends Model {
...
#Reference
#Indexed
public SiteUser commenter;
public static List<Comment> getLastCommentsByUsers(final List<SiteUser> users) {
final Query<Comment> query ds().createQuery(Comment.class);
query.field(commenter).hasAnyOf(users);
return query.asList();
}
}
SiteUser:
#Entity(noClassnameStored=true)
public class SiteUser extends AbstractUser {
public String realName;
}
AbstractUser:
public class AbstractUser extends Model {
#Indexed(value= IndexDirection.DESC, unique = true)
public String emailAddress;
#Required
public String password;
}
The method getLastCommentsByUsers() is supposed to return all comments by the users in the users parameter, but I always get an empty List back. The reason that Commment is a separate collection is to be able to retrieve last X Comments by certain users across their associated Posts, which isn't possible if the Comment is embedded in the Post collection.
Is there something wrong with my query (should I be using something other than hasAnyOf), or is it a problem with the relationship mapping - should I be using ObjectId instead?
I use the in() method with a list or set and its working perfectly. Here's a snippet:
List<String> keywordList;
List<Product> products = Product.find().field("keywords").in(keywordList).asList();
This should work for collection of embedded or references too.
You should use List<Key<SiteUser>> to query:
public static List<Comment> getLastCommentsByUsers(final List<SiteUser> users) {
final Query<Comment> query ds().createQuery(Comment.class);
query.field(commenter).hasAnyOf(toKeys(users)); // convert to keys
return query.asList();
}
public static List<Key<SiteUser>> toKeys(List<SiteUser> users) {
List<Key<SiteUser>> keys = new ArrayList<Key<SiteUser>>();
for(SiteUser user: users) {
keys.add(ds().getMapper().getKey(user));
}
return keys;
}
Or you can just get the keys by:
List<Key<SiteUser>> keys = ds().createQuery(SiteUser.class).query().filter(...).asKeyList();