Adrian Matei (aka adixchen) Reading time ~1 minute

With Java version 8 it’s very easy to get the current timestamp with LocalDateTime.now(). One problem that might arise is it’s not easily testable/mockable. Imagine some testing data and conditions depend on the date you set through this call, or you want to do “time travel”. The solution is quite simple - define a service class that produces the current timestamp, which you can then easily mock, in this case we’ll name it ProductionCalendar:

@ApplicationScoped
public class ProductionCalendar {
  public LocalDateTime currentTimestamp() {
    return LocalDateTime.now();
  }

  public LocalDate currentDate() {
    return LocalDate.now();
  }

  public String currentTimestamp(Partner partner) {
    return DateTimeFormatter.ofPattern("yyyy-MM-dd-HH.mm.ss.SSSSSS")
        .format(LocalDateTime.now());
  }
}

Then when were you need to call LocalDateTime.now() inject this service and use its methods instead:

@Stateless
public class PartnerRepository {

  @Inject ProductionCalendar productionCalendar;

  @Inject EntityManager entityManager;

  public void savePartner(String partnerNumber) {
    Partner partner = new Partner();
    partner.setPartnerNumber(partnerNumber);
    partner.setCreatedAt(productionCalendar.currentTimestamp());

    entityManager.persist(partner);
  }

  public List<Partner> getPartnerCreatedAfterDate(LocalDateTime timestamp) {
    String queryJpql = "select p from P p where createdAt > :timestamp order by createdAt desc";
    TypedQuery<Partner> query = entityManager.createQuery(queryJpql, Partner.class);
    query.setParameter("timestamp", timestamp);

    return query.getResultList();
  }
}

Then, in the test class you can easily mock ProductionCalendar and calls to its methods and set timestamps as we need them:

@ExtendWith(MockitoExtension.class)
@RunWith(JUnitPlatform.class)
public class UserServiceUnitTest {

  @Inject PartnerRepository partnerRepository;

  @Mock ProductionCalendar productionCalendar;

  @Test
  void testGetLatestPartners() {
     when(productionCalendar.currentTimestamp()).thenReturn(LocalDatetime.now().minusDays(1));
     partnerRepository.save("12345");
     var recentPartners = partnerRepository.getPartnerCreatedAfterDate(LocalDatetime.now().minusDays(2));
     assertThat(recentPartners, hasSize(1));
  }

}

Shared with from Codever. 👉 Use the Copy to mine functionality to copy this snippet to your own personal collection and easy manage your code snippets.

Codever is open source on Github ⭐🙏

Adrian Matei (aka adixchen) Reading time ~1 minute

In the snippet you can see a message driven bean set up to consume asynchronously messages from a JMS queue:

import javax.ejb.ActivationConfigProperty;
import javax.ejb.MessageDriven;
import javax.ejb.MessageDrivenContext;
import javax.ejb.TransactionAttribute;
import javax.ejb.TransactionAttributeType;
import javax.inject.Inject;
import javax.jms.JMSException;
import javax.jms.Message;
import javax.jms.MessageListener;
import javax.jms.TextMessage;
import javax.xml.bind.JAXBException;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

@MessageDriven(
    name = "AccountMessageBean",
    activationConfig = {
      @ActivationConfigProperty(
          propertyName = "destinationType",
          propertyValue = "javax.jms.Queue"),
      @ActivationConfigProperty(
          propertyName = "destination",
          propertyValue = "java:jboss/exported/queue/account.queue"),
      @ActivationConfigProperty(
          propertyName = "acknowledgeMode",
          propertyValue = "Auto-acknowledge")
    })
public class AccountMessageBean implements MessageListener {

  private static final Logger logger = LoggerFactory.getLogger(AccountMessageBean.class);

  @Inject  MessageDrivenContext messageDrivenContext;

  @Inject MessageService messageService;

  @Override
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  public void onMessage(Message message) {
    if (!(message instanceof TextMessage)) {
      logger.warn("This is not a text message. Break");
      return;
    }

    TextMessage messageText = (TextMessage) message;
    try {
      if (messageText.getText().contains("organisation")) {
        this.messageService.createOrganisation(messageText.getText());
      } else if (messageText.getText().contains("person"))
        this.messageService.createPerson(messageText.getText());
      else {
        logger.warn("Something went wrong, must be one of those types");
      }

    } catch (JMSException | JAXBException e) {
      logger.error("An error occured while processing the message, mark to rollback");
      messageDrivenContext.setRollbackOnly();
    }
  }
}

Note the following:

  • @MessageDriven annotation where the type of the consumed resource is defined (destinationType - here javax.jms.Queue, but it could also be javax.jms.Topic), the location of the queue (propertyName = "destination") and that the message from the queue is acknowledged once the processing is successful (this is the default setup)
  • the consuming message driven bean implements the javax.jms.MessageListener and the method onMessage(Message message)
  • if a container-managed transaction is not present a new one is created - @TransactionAttribute(TransactionAttributeType.REQUIRED)
  • the consumed message is expected to be text (javax.jms.TextMessage)
  • you access the string value of the message by calling the getText() method - messageText.getText()
  • MessageDrivenContext is used to gain additional access to transaction management - here the transaction is marked for rollback, if an error occurs messageDrivenContext.setRollbackOnly(); (this is only possible in container-based transactions)

Reference - https://docs.oracle.com/javaee/6/tutorial/doc/bncgl.html#bncgq

Shared with from Codever. 👉 Use the Copy to mine functionality to copy this snippet to your own personal collection and easy manage your code snippets.

Codever is open source on Github ⭐🙏

Adrian Matei (aka adixchen) Reading time ~1 minute

Use the firstResult(int startPosition) and setMaxResults(int maxResult) of the JPA Query interface. The firstResult() method sets the offset value in SQL lexicon, meaning the position of the first result to retrieve. The setMaxResults() method sets the limit value in SQL lexicon, meaning the maximum number of results to retrieve:

@Stateless
public class PartnerInitialLoadRepository {

  @Inject private EntityManager em;

  public List<PartnerInitialLoad> getNextUnprocessed(Integer pageSize, Integer offset) {
    var jpqlQuery="select p from PartnerInitialLoad p where p.status = 'UNPROCESSED' order by p.partnernumber asc";
    var query =
        em.createNamedQuery(jpqlQuery, PartnerInitialLoad.class);
    query.setFirstResult(offset);
    query.setMaxResults(pageSize);

    return query.getResultList();
  }
}

JPA with Hibernate translates this in the following SQL statement in Oracle dialect which uses offset and limit:

select partnerini0_.PARTNER_NUMBER as PARTNER_1_13_,
       partnerini0_.COMPLETED_AT as COMPLETE2_13_,
       partnerini0_.STATUS as STATUS3_13_
from T_PARTNER_INITIAL_LOAD partnerini0_
where partnerini0_.STATUS is null
   or partnerini0_.STATUS = 'UNPROCESSED'
order by partnerini0_.PARTNER_NUMBER asc
offset ? limit ?;

An alternative would be to use native queries, the major drawback is that you make your code dependent on the underlying RDBMS:

  public List<PartnerInitialLoad> getNextUnprocessed(Integer pageSize, Integer offset) {
    var sqlNativeString =
        String.format(
            "select * from T_PARTNER_INITIAL_LOAD where STATUS = 'UNPROCESSED' order by PARTNER_NUMBER asc OFFSET %s LIMIT %s",
            offset, pageSize);
    var query = em.createNativeQuery(sqlNativeString, PartnerInitialLoad.class);

    return query.getResultList();
  }

Shared with from Codever. 👉 Use the Copy to mine functionality to copy this snippet to your own personal collection and easy manage your code snippets.

Codever is open source on Github ⭐🙏

Adrian Matei (aka adixchen) Reading time ~1 minute

To delete one entry you can use findOneAndRemove command - it issues a mongodb findAndModify remove command. Finds a matching document, removes it, passing the found document (if any) to the callback.

let deleteBookmarkById = async (userId, bookmarkId) => {
  const bookmark = await Bookmark.findOneAndRemove({
    _id: bookmarkId,
    userId: userId
  });

  if ( !bookmark ) {
    throw new NotFoundError('Bookmark NOT_FOUND with id: ' + bookmarkId);
  } else {
    return true;
  }
};

An alternative is to use the deleteOne() method which deletes the first document that matches conditions from the collection. It returns an object with the property deletedCount indicating how many documents were deleted:

let deleteBookmarkById = async (userId, bookmarkId) => {
  const response = await Bookmark.deleteOne({
    _id: bookmarkId,
    userId: userId
  });

  if ( response.deletedCount !== 1 ) {
    throw new NotFoundError('Bookmark NOT_FOUND with id: ' + bookmarkId);
  } else {
    return true;
  }
};

To delete multiple documents use the deleteMany function. This deletes all the documents that match the conditions specified in filter. It returns an object with the property deletedCount containing the number of documents deleted.

/**
 * Delete bookmarks of a user, identified by userId
 */
let deleteBookmarksByUserId = async (userId) => {
  await Bookmark.deleteMany({userId: userId});
  return true;
};

Reference - https://mongoosejs.com/docs/api/model.html

Shared with from Codever. 👉 Use the Copy to mine functionality to copy this snippet to your own personal collection and easy manage your code snippets.

Codever is open source on Github ⭐🙏

Adrian Matei (aka adixchen) Reading time ~1 minute

You can define several named queries on an entity by using the @NamedQueries and @NamedQuery annotations. For each named query you need to define a name and the jpql query itself:

import javax.persistence.*;

@Entity
@Access(AccessType.FIELD)
@Table(name = PartnerInitialLoad.TABLE_NAME)
@NamedQueries({
  @NamedQuery(
      name = FIND_MAX_PARTNERNUMMER,
      query = "select max(p.partnernummer) from PartnerInitialLoad p"),
  @NamedQuery(
      name = FIND_PARTNER_BY_STATUS,
      query =
          "select p from PartnerInitialLoad p where p.status = :status order by p.partnernummer asc")
})
public class PartnerInitialLoad {
  public static final String TABLE_NAME = "T_PARTNER_INITIAL_LOAD";

  public static final String FIND_MAX_PARTNERNUMMER = "findMaxPartnernummer";
  public static final String FIND_PARTNER_BY_STATUS = "findPartnerByStatus";

  public static final String PARAM_STATUS = "status"; //the value here has to match the one in jpql, here "status"

 // further entity details emitted for brevity
}

Then use the named queries in your repository services. For that use the createNamedQuery method of the EntityManager, which expects the name of the query you defined in the metadata, plus the type of the query result:

@Stateless
public class PartnerInitialLoadRepository {

  @Inject private EntityManager em;

  public List<PartnerInitialLoad> getPartnersByStatus(Integer chunkSize, String status) {
    var query =
        em.createNamedQuery(PartnerInitialLoad.FIND_UNPROCESSED_PARTNER, PartnerInitialLoad.class);
    query.setParameter(PartnerInitialLoad.PARAM_STATUS, status);
    query.setMaxResults(chunkSize);

    return query.getResultList();
  }

  public int getMaxPartnernummer() {
    var query = em.createNamedQuery(PartnerInitialLoad.FIND_MAX_PARTNERNUMMER, Integer.class);
    var singleResult = query.getSingleResult();

    return singleResult == null ? 0 : singleResult;
  }
}

Shared with from Codever. 👉 Use the Copy to mine functionality to copy this snippet to your own personal collection and easy manage your code snippets.

Codever is open source on Github ⭐🙏