jlcanela / play-slick Goto Github PK

#About

This plugin makes Slick a first-class citizen of Play 2.1.

The play-slick plugins consists of 2 parts:

• DDL schema generation Plugin that works like the Ebean DDL Plugin. Based on config it generates create schema and drop schema SQL commands and writes them to evolutions.
• A wrapper DB object that uses the datasources defined in the Play config files. It is there so it is possible to use Slick sessions in the same fashion as you would Anorm JDBC connections.

The intent is to get this plugin into Play 2.2 if possible.

#Usage In the project/Build.scala file add::

.dependsOn(RootProject( uri("git://github.com/freekh/play-slick.git") ))

to your play.Project

Example::

val main = play.Project(appName, appVersion, appDependencies).settings(
  // Add your own project settings here      
 ).dependsOn(RootProject( uri("git://github.com/freekh/play-slick.git") ))

##DDL plugin

In order to enable DDL schema generation you must specify the packages or classes you want to have in the application.conf file: slick.default="models.*" It follows the same format as the Ebean plugin: slick.default="models.*" means all Tables in the models package should be run on the default database.

It is possible to specify individual objects like: slick.default="models.Users,models.Settings"

##DB wrapper

The DB wrapper is just a thin wrapper that uses Slicks Database classes with databases in the Play Application .

This is an example usage:

import play.api.db.slick.Config.driver.simple._

play.api.db.slick.DB.withSession{ implicit session =>
  Users.insert(User("fredrik","ekholdt"))
}

Or transactionally:

import play.api.db.slick.Config.driver.simple._

play.api.db.slick.DB.withTransaction{ implicit session =>
  val list = Query(Users).filter(name === "fredrik")
  val updatedUsers = update(list)
  Users.insertAll(updatedUsers)
}

You can load a different datasource than the default (defined by db.default.* in application.conf), using the DB name parameter :

play.api.db.slick.DB("myOtherDb").withSession{ implicit session =>
  Users.insert(User("fredrik","ekholdt"))
}

Using import play.api.db.slick.Config.driver.simple._ will import the driver defing with the key db.default.driver in application.conf, or the one set by the test helpers in test mode (see test section for more information).

If you need to use more than one driver per mode, you can read the next section!

##Multiple datasources and drivers

When having multiple datasources and drivers it is recommended to use the cake pattern. Do not worry about the scary name it is quite easy.

Hava look in the samples for an example of this or keep reading.

For each table you have, create a self-type of play.api.db.slick.Profile and import everything from the profile on your table:

trait UserComponent extends Profile { this: Profile =>
   import profile.simple._

   object Users extends Table[User]("USERS") { ... }
}

Then you just have to put all your tables together into a DAO (data access object) like this:

class DAO(override val profile: ExtendedProfile) extends UserComponent with FooComponent with BarComponent with Profile

Whenever you need to use you database you just new up your DAO and import everything in it and in its profile:

package models

import play.api.slick.DB
object current {
  val db = DB("mydb")
  val dao = new DAO(db.driver(play.api.Play.current))      
} 

or if you just want the default database :

val dao = new DAO(DB.driver(play.api.Play.current))

In your application.conf you can now do: slick.default="models.current.*"

You can then use it like this:

import current._
import current.simple._

DB.withSession{ implicit session => 
   Query(Users).list
}

If needed, you can also use another driver like this :

object specificDb {
 val dao = new DAO(scala.slick.driver.MySQLDriver)     
}

Pweeh, there are a certain amount of lines of code there, but works great and scales along with the life cycle of your app: from the start, when you need tests, when you change the DB, ...

##Writing tests

To use a test datasource, you can use the inMemoryDatabase helper :

import play.api.test.Helpers._

class DBSpec extends Specification {

  "DB" should {

   "my test" in new WithApplication(FakeApplication(additionalConfiguration = inMemoryDatabase())) {
        play.api.db.slick.DB.withSession{ implicit session =>
          //...
      }
    }

  }
}

This will use a H2 datasource with this url : "jdbc:h2:mem:play-test"

Off course, you can also use the default datasource :

"default ds" in new WithApplication {
    play.api.db.slick.DB.withSession{ implicit session =>
      //
}

Or a specific one :

"specific ds" in new WithApplication {
  play.api.db.slick.DB("specific").withSession{ implicit session =>
    //
}

Copyright: Typesafe 2013 License: Apache License 2.0, http://www.apache.org/licenses/LICENSE-2.0.html