A robust, powerful, and very simple ORM android database library with annotation processing.

The library eliminates the need for writing most SQL statements, writing ContentValues for every table, converting cursors into models, and so much more.

Let DBFlow make SQL code flow like a steady stream so you can focus on your complex problem and not be hindered by repetitive code writing.

This library is based on Active Android, Schematic, Ollie, and Sprinkles, but takes the best of each while offering much more functionality and extensibility.

What sets this library apart: every feature has been unit tested to ensure functionality, baked in support for multiple databases seamlessly, powerful and fluid builder logic in expressing SQL statements, annotation processing to enable blistering speed, ModelContainer classes that enable direct to database parsing for data such as JSON, and rich interface classes that enable powerful flexibility.

1. Added SQLite Triggers!
2. Added the IN and NOT IN operators to Condition
3. Fixes an issue where a FlowCursorList was A. not caching models and B. not handling ModelView properly, thanks Cain.
4. Added ForeignKeyAction which specify the action to take when updates and deletes occur to the foreign key. Thanks Michal
5. Added the Insert statement wrapper! Particularily useful for Trigger

for older changes, from other xx.xx versions, check it out here

Add the maven repo url to your build.gradle:

  repositories {
        maven { url "https://raw.github.com/Raizlabs/maven-releases/master/releases" }
  }

Add the library to the project-level build.gradle, using the apt plugin and the AARLinkSources plugin::

  dependencies {
    apt 'com.raizlabs.android:DBFlow-Compiler:1.2.0'
    aarLinkSources 'com.raizlabs.android:DBFlow-Compiler:1.2.0:sources@jar'
    compile 'com.raizlabs.android:DBFlow-Core:1.2.0'
    aarLinkSources 'com.raizlabs.android:DBFlow-Core:1.2.0:sources@jar'
    compile 'com.raizlabs.android:DBFlow:1.2.0'
    aarLinkSources 'com.raizlabs.android:DBFlow:1.2.0:sources@jar'
  }

Not supported as google is no longer supporting it.

We need to configure the FlowManager properly. Instead of passing in a Context wherever it is used, we hold onto the Application context instead and reference it.

You will need to extend the Application class for proper configuration:

public class ExampleApplication extends Application {

    @Override
    public void onCreate() {
        super.onCreate();
        FlowManager.init(this);
    }

    @Override
    public void onTerminate() {
        super.onTerminate();
        FlowManager.destroy();
    }
}

Lastly, add the definition to the manifest (with the name that you chose for your custom application):

<application
  android:name="{packageName}.ExampleApplication"
  ...>
</application>

First class you need to define is the @Database. It is recommended you store the name and version as static final fields. The database name is not required for singular databases, however it is good practice to include it here.

@Database(name = AppDatabase.NAME, version = AppDatabase.VERSION, foreignKeysSupported = true)
public class AppDatabase {

    public static final String NAME = "App";

    public static final int VERSION = 1;
}

Second, you need to define at least one @Table class. The databaseName is only required when dealing with multiple databases. You can either implement the Model interface or extend BaseModel.

@Table(databaseName = TestDatabase.NAME)
public class TestModel1 extends BaseModel {
    @Column(columnType = Column.PRIMARY_KEY)
    public
    String name;
}

So you have an existing DB you wish to include in your project. Just name the database the same as the database to copy, and put it in the app/src/main/assets/ directory.

in DBFlow migrations are separate, public classes that contain both the @Migration and Migrationinterface. If you are using multiple databases, you're required to specify it for the migration.

@Migration(version = 2, databaseName = TestDatabase.NAME)
public class Migration1 extends BaseMigration {

    @Override
    public void onPreMigrate() {
      // called before migration, instantiate any migration query here
    }

    @Override
    public void migrate(SQLiteDatabase database) {
      // call your migration query
    }

    @Override
    public void onPostMigrate() {
      // release migration resources here
    }
}

The SQL language is wrapped in a nice builder notation. DBFlow generates a $Table containing static final column strings to use in your queries!

new Select().from(DeviceObject.class)
                             .where(Condition.column(DeviceObject$Table.NAME).is("Samsung-Galaxy S5"))
                             .and(Condition.column(DeviceObject$Table.CARRIER).is("T-MOBILE"))
                             .and(Condition.column(DeviceObject$Table.LOCATION).is(location);

To see more go to the full tutorial

Model containers are classes that imitate and use the blueprint of Model classes in order to save data such as JSON, Hashmap, or your own kind of data to the database. To create your own, extend the BaseModelContainer class or implement the ModelContainer interface. More info here

For example here is the JSONModel implementation:

public class JSONModel<ModelClass extends Model> extends BaseModelContainer<ModelClass, JSONObject> implements Model {

    public JSONModel(JSONObject jsonObject, Class<ModelClass> table) {
        super(table, jsonObject);
    }

    public JSONModel(Class<ModelClass> table) {
        super(table, new JSONObject());
    }

    @Override
    @SuppressWarnings("unchecked")
    public BaseModelContainer getInstance(Object inValue, Class<? extends Model> columnClass) {
        return new JSONModel((JSONObject) inValue, columnClass);
    }

    @Override
    public JSONObject newDataInstance() {
        return new JSONObject();
    }

    @Override
    public Object getValue(String columnName) {
        return getData().opt(columnName);
    }

    @Override
    public void put(String columnName, Object value) {
        try {
            getData().put(columnName, value);
        } catch (JSONException e) {
            FlowLog.logError(e);
        }
    }
}

And then in every Model class you wish to use this class, you need to add the annotation @ContainerAdapter. This generates the definition required to save objects correctly to the DB.

TypeConverter allows non-Model objects to save to the database by converting it from its Model value to its Database value. These are statically allocated accross all databases. More info here

@com.raizlabs.android.dbflow.annotation.TypeConverter
public class CalendarConverter extends TypeConverter<Long, Calendar> {

    @Override
    public Long getDBValue(Calendar model) {
        return model.getTimeInMillis();
    }

    @Override
    public Calendar getModelValue(Long data) {
        Calendar calendar = Calendar.getInstance();
        calendar.setTimeInMillis(data);
        return calendar;
    }
}

ModelView are a special kind of Model that creates a database VIEW based on a special SQL statement. They must reference another Model class currently.

@ModelView(query = "SELECT * FROM TestModel2 WHERE model_order > 5", databaseName = TestDatabase.NAME)
public class TestModelView extends BaseModelView<TestModel2> {
    @Column
    long model_order;
}

Trigger are actions that are automatically performed before or after some action on the database. For example, we want to log changes for all updates to the name on the Friend table.

CompletedTrigger<Friend> trigger = new Trigger<Friend>("NameTrigger")
                                    .after().update(Friend.class, Friend$Table.NAME)
                                    .begin(
                                        new Insert<FriendLog>(FriendLog.class)
                                          .columns(FriendLog$Table.OLDNAME, FriendLog$Table.NEWNAME, FriendLog$Table.DATE)
                                          .values("old.Name", "new.Name", System.currentTimeMillis())
                                          };