Awesome
Tiny Generic Dao in Java
It's an open source (Apache License) tiny generic dao in Java, with minimal dependencies(only slf4j-api for dada-core) and a quick learning curve. The goal of this project is to create a tiny generic dao in Java with a micro core and multiple independent implementations (inmemory, csv, hibernate, jpa, iciql).
This project could be in time a Spring Data without Spring. :smile:
The project is in early stage but it's functional. I will try to add more documentation, javadoc, unit tests.
Any contribution is very welcome. I hope to create a community around this project.
Modules
Practically Dada is a microframework and the aim is to keep the core simple but extensible. I try to create a little ecosystem (extensions/modules) based on this core with the help of the community. For now are available these modules:
- csv (very useful for tests and fast prototypes)
- iciql
Maybe the modules for JPA, Hibernate are the next implementations.
Components
- GenericDao is the base interface for all Dao.
- InMemoryGenericDao is a bonus implementation. Can be used with success in tests and prototypes.
- EntityDao is a simple convenient wrapper over GenericDao that works with Entity classes.
- GenericService, EntityService are bonus implementations for people that are using three tier architecture (Dao <-> Service).
Using Maven
In your pom.xml you must define the dependencies to Dada artifacts with:
<dependency>
<groupId>ro.fortsoft.dada</groupId>
<artifactId>dada-core</artifactId>
<version>${dada.version}</version>
</dependency>
where ${dada.version} is the last dada version.
You may want to check for the latest released version using Maven Search
How to use
It's very simple to add dada in your application:
public static void main(String[] args) {
CompanyDao companyDao = new MyDaoFactory().createCompanyDao();
System.out.println(">>> Find company with id " + 1L);
Company company = companyDao.findById(1L);
System.out.println("company = " + company);
System.out.println(">>> Find all companies");
List<Company> companies = companyDao.findAll();
System.out.println("companies = " + companies);
System.out.println(">>> Delete company with id " + 1L);
companyDao.deleteById(1L);
System.out.println(">>> Find all companies");
companies = companyDao.findAll();
System.out.println("companies = " + companies);
}
where Company
is a simple entity:
public class Company extends AbstractEntity {
private String name;
private String fiscalCode; // CUI or CIF
public Company() {
}
public Company(Long id) {
super(id);
}
public String getName() {
return name;
}
public Company setName(String name) {
this.name = name;
return this;
}
public String getFiscalCode() {
return fiscalCode;
}
public Company setFiscalCode(String fiscalCode) {
this.fiscalCode = fiscalCode;
return this;
}
@Override
public String toString() {
return "Company{" +
"id=" + id +
", name='" + name + '\'' +
", fiscalCode='" + fiscalCode + '\'' +
'}';
}
}
I extended AbstractEntity
because I want to bypass to this class some boring stuff (set/get id, hashCode, equals methods) but it's not mandatory to extend it.
The CompanyDao
interface looks like:
public interface CompanyDao extends EntityDao<Company> {
// define bellow non crud (business specific) methods
public Company findByFiscalCode(String fiscalCode);
}
A possible implementation (I am using Iciql implementation) for CompanyDao
can be:
public class IciqlCompanyDao extends IciqlDao<Company> implements CompanyDao {
public IciqlCompanyDao() {
super();
}
public IciqlCompanyDao(IciqlDbFactory dbFactory) {
super(dbFactory);
}
@Override
public Company findByFiscalCode(String fiscalCode) {
Company alias = getAlias();
return getDb().from(alias).where(alias.getFiscalCode()).is(fiscalCode).selectFirst();
}
}
Demo
I have a tiny demo application. The demo application is in dada-demo folder.
How to build
Requirements:
- Git
- JDK 7 (test with
java -version
) - Apache Maven 3 (test with
mvn -version
)
Steps:
- create a local clone of this repository (with
git clone https://github.com/decebals/dada.git
) - go to project's folder (with
cd dada
) - build the artifacts (with
mvn clean package
ormvn clean install
)
After above steps a folder dada/target is created and all goodies are in that folder.
Versioning
Dada will be maintained under the Semantic Versioning guidelines as much as possible.
Releases will be numbered with the follow format:
<major>.<minor>.<patch>
And constructed with the following guidelines:
- Breaking backward compatibility bumps the major
- New additions without breaking backward compatibility bumps the minor
- Bug fixes and misc changes bump the patch
For more information on SemVer, please visit http://semver.org.