0
votes

I have a model that is composed of three case classes as below:

case class MyModel(myId: MyIdType, name: MyNameType)

case class MyIdType(id: Long)
case class MyNameType(name: String)
object MyNameType(name: String) {
  val NAME1 = MyNameType("name1")
  val NAME2 = MyNameType("name2")
  val NAME3 = MyNameType("name3")
}

Let's say these are the existing models. I have a Slick table mapping that goes like this:

  class MyTable(tag: Tag) extends Table[MyTableElem](tag, "myTable") {
    def id                  = column[Long]("id", O.PrimaryKey)
    def name                = column[String]("name")

    def * = (id, name) <> (MyTableElem.tupled, MyTableElem.unapply)
  }

As it can be seen that I have to have another type to map my table MyTable first to MyTableElem and then during every read, I transform the MyTableElem to MyModel. Is there a way to avoid this and go directly to the MyModel? I guess I have to implement the tupled and unapply methods or?

1

1 Answers

1
votes

I think you may have to define your own column type. In Slick 3.0 documentation Using Custom Scalar Types in Queries, MappedTo could be used in your case.

If you have a wrapper class (which can optionally be a case class and/or value class) for an underlying value of some supported type, you can make it extend MappedTo to get a macro-generated implicit ColumnType for free.

Following is how I try to get your problem done.

import slick.lifted.Tag
// I am using PostgreSQL
import slick.driver.PostgresDriver.api._

case class MyModel(myId: MyIdType, name: MyNameType)

case class MyIdType(id: Long) extends MappedTo[Long] {
  override def value: Long = id
}
case class MyNameType(name: String) extends MappedTo[String] {
  override def value: String = name
}

class MyTable(tag: Tag) extends Table[MyModel](tag, "myTable") {
  def id                  = column[MyIdType]("id", O.PrimaryKey)
  def name                = column[MyNameType]("name")

  def * = (id, name) <> (MyModel.tupled, MyModel.unapply)
}

Anyway, hope it helps.