I am working in a project which uses JPA ORM and framework provides two kinds of method to create queries.
entityManager.createQuery(query1);
entityManager.createNativeQuery(query2);
I understand the kinds of query string is to be passed to use them, but I don't know exactly why do we need to create native query? Probably we don't want to use ORM capabilities there?
You do not need to create a native query unless you want to. JPQL eventually is translated into SQL by the framework but the framework lets you call the native query also. Why would want to do that:
Low level access, which means that you can optimize and handle the mapping by yourself; with SQL you actually access the database table while with JPQL you access the entity objects;
Maybe you do not want to learn JPQL if you already know SQL
You already have the queries written in SQL, and do not have resources/time to port them to JPQL
createQuery uses JPAs own query language, you select from Class names instead of table names. This is not SQL, it is just similar, and is later transformed to real SQL. Mapping to java classes will be done automatically and actual class instances will be returned as result.
createNativeQuery uses real SQL, and will not be able to use JPA features. This method is used in general if you need to do something really odd that is not supported by JPA. A list of Object[] will be returned, and mapping to java objects will have to be done manually. In other words, its just like working with a DB before JPA came to, just slightly more convenient since connection handling is done automatically.
I have used it for optimization purposes. Using Native queries means that the ORM mapping is not in place, and instead of JPQL, you use the DB's native syntax. So, as #RasmusFranke also pointed out, if you need something that is not supported by JPA (like when you want to use DB vendor specific extensions, which is conceptually a bad idea, since JPA is all about being DB agnostic, but happens nevertheless. I know...)
The other effect of this is that by using native queries, only the supplied query is run. No eager fetching of other entities, or other unwanted stuff. This way, if you deal with huge amounts of objects, you can save some heap space.
Related
After making some search on the web, I think that when using Spring Data JPA Named Queries, we need some extra implementation or definitions comparing to the derived or dynamic queries in Spring Data JPA. In this scene, I am really wondering that do we really need to use Spring Data JPA Named Queries?
Spring Data derived queries are intended (and useful) only for very simple queries. Those queries where you look at the name that you would naturally give such a method and would immediately know how to implement it in SQL or JPQL.
As soon as a query gets a little more complex we shouldn't use derived queries anymore, and often we can't even if we wanted to. For example query derivation doesn't have a way to control the precedence between AND and OR.
For all other queries we need to explicitly code the query one way or the others. And if you don't want your queries mixed with your repository, a named query is a very viable alternative.
We have a (possibly large) custom data structure implemented in Java (8+). It has a simple and optimal API for querying pieces of data. The logical structure is roughly similar to an RDMS (it has e. g. relations, columns, primary keys, and foreign keys), but there is no SQL driver.
The main goal is to access the data via ORM (mapping logical entities to JPA annotated beans). It would be nice if we could use JPQL. Hibernate is preferred but other alternatives are welcome too.
What is the simplest way to achieve this? Which are the key parts of such an implementation?
(P. S. Directly implementing SessionImplementor, EntityManagerImplementor etc. seems to be too complicated.)
You have two possibilities.
Implement a JDBC compliant driver for your system, so you can use a JPA implementation such as Hibernate "directly" (although you may need to create a custom dialect for your system).
Program directly against the JPA specification like ObjectDB does, which bypasses the need to go through SQL and JPA implementations completely.
The latter one is probably easier, but you'd still need to implement the full JPA API. If it's a custom in-house-only system, there's very little sense in doing either one.
One idea I thought up just now, that I feel may work is this:
Use an existing database implementation like H2 and use the JPA integration with that. H2 already has a JPA integration libraries, so it should be easy.
In this database, create a Java stored procedure or function and call it from your current application through JPA. See this H2 documentation on how to create a Java stored procedure or function. (You may want to explore the section "Using a Function as a Table" also.)
Define a protocol for the service methods and encapsulate it in a model class. An instance of this model class may be passed to the function/SP and responses retrieved.
Caveat: I have never done this myself but I think it will work.
Edit: Here is a diagram representing the thought. Though the diagram show H2 separately, it will most probably be in the same JVM as "Your Java/JEE application". However, since it is not necessary to use H2, I have shown it as as separate entity.
While joining multiple tables on my project using Hibernate jpa /Spring (annotation driven), I had to use the NamedNativeQuery annotation to achieve my objective to extract a distributed resultset spanning multiple tables. This may be a question that merely serves academic merit, but given that I am starting out on Hibernate - is there another way to achieve table joins without having to fall back on queries native to the database dialect?
Yes. I believe this is exactly what you need: https://docs.jboss.org/hibernate/entitymanager/3.5/reference/en/html/querycriteria.html#querycriteria-tuple
Criteria Queries is a way of building a complete query just using its API. If I were you, I'd give it a try.
By the way, according to your question, the reason for using native queries is just for retrieving a specific set of columns. If this is the case, you can also write it using HQL as well. The query doesn't necessarily needs to be native.
How many ways Hibernate provide to access database?
For example, I want to CRUD an object to database, I found out:
Using session from SessionFactory:
session.save(object);
...
Using Hibernate Query Language.
Using Hibernate Criteria Queries.
Using Native SQL.
But I don't know what I should use. Please list your practice to access database in PRIORITY DECREASING ORDER and the reason why you do that.
Thank you.
If you have an ID and wants the associated entity, the use Session.get(). It's efficient, and makes use of the first-level cache to avoid reexecuting the query again and again.
If you need to get entities via other criteria (like all the users with a given first name, for example), then use JPQL queries. They are simple to write, very readable, and have less limitations than criteria queries.
If you need to take various optional criteria (like for a complex search form), the criteria API is the tool for the job. But it can't do everything a JPQL query does. There are other APIs available, and you can relatively easily write an API that generates dynamic JPQL queries if needed.
If you have a really complex query that can't be expressed using JPQL, then use SQL.
To write things to the database, queries should generally not be used, except in very specific circumstances where many entities must be modified the same way. Instead, get the entities to modify, and modify them. Hibernate will save their new state automatically.
Is there any benefit of using #NamedQuery over #NamedNativeQuery in hibernate or vice verse.
I cant spot the exact difference or
in which context we should use #NamedQuery over #NamedNativeQuery
Thanks in advance.
#NamedNativeQuery lets you write a named SQL query, while #NamedQuery lets you write a named HQL query (or JPQL).
In general, you should prefer to write HQL queries because then you can let Hibernate handle the intricacies of converting the HQL into the various SQL dialects. This will make your job much simpler when you choose to switch DBMS providers.
When taking about performance, you have to know something about what goes on under the hood.
You have probably programmed something using straightforward JDBC, so you know how to queries
get passed to the driver and send to the database. When using HQL or JPA-QL the queries first have
to be parsed into an SQL language that the database can understand. In this case, we have an extra
parsing step in between. Note that Native SQL queries, including stored procedure calls, the persistence
framework still takes care of mapping the JDBC result sets to graphs of persistent objects.
If you want to include a native SQL hint to instruct the database management systems query optimizer,
for example, you need to write the SQL yourself. HQL and JPA-QL do not have keywords for this.
The disadvantage of putting native SQL in your mapping metadata is lost database portability, because
your mappings, and hence your application, will work only for a particular database. But usually this is of a
minor concern as you are probably not creating a framework that has to work on every database.
When you want to get behind the performance of your query, you really have to consult the database
and look at the execution plan - A DBA can tell you exactly what is good and what can be optimized.
#NamedQuery should be constructed with the query language (HQL or persistence query language). #NamedNativeQuery should be constructed with native SQL.