I have a Scala compile error I have been unable to find any information for. I am using slick 3.0 and am getting a compile error of
value ~ is not a member of slick.lifted.Rep[Option[Int]]
I believe the issue relates to the way I am using an Option to represent my ID field.
I have tried adding id.?
to the id field as suggested in this answer but I still get a compilation error in the same vein. Has anything change in slick 3.0?
My code is as follows:
import slick.driver.H2Driver.api._
import scala.concurrent.ExecutionContext.Implicits.global
case class Recipe(id: Option[Int] = None, name: String, instructions: String, ingredients: String)
object AddFixtures {
class Recipes(tag: Tag) extends Table[Recipe](tag, "recipe") {
def id = column[Int]("id", O.PrimaryKey, O.AutoInc)
def name = column[String]("name")
def instructions = column[String]("instructions")
def ingredients = column[String]("ingredients")
def * = id ~ name ~ instructions ~ ingredients <> (Recipe, Recipe.unapply _)
}
val recipes = TableQuery[Recipes]
val setup = DBIO.seq(
recipes.schema.create,
recipes += Recipe(None, "Chicken with Avocado", "Mix it up", "Chicken, Avocado")
)
def apply() = {
val db = Database.forConfig("h2mem1")
try db.run(setup)
finally db.close
}
}
I think that the problem is that some symbols are not used like before in slick 3.0.0
Take a look here for further problems
in you case the problematic line will be something like this depending what are you going to do, but this should work:
def * = (id, name, instructions, ingredients) <> ((Recipe.apply _).tupled, Recipe.unapply _)
Also you do not need the implicits import
and there is also a problem with your option[Int] parameter: maybe this should be better:
import slick.driver.H2Driver.api._
object SlickStackOverflow extends App {
}
case class Recipe(id: Option[Int] = None, name: String, instructions: String, ingredients: String)
object AddFixtures {
class Recipes(tag: Tag) extends Table[Recipe](tag, "recipe") {
def id = column[Option[Int]]("id", O.PrimaryKey, O.AutoInc)
def name = column[String]("name")
def instructions = column[String]("instructions")
def ingredients = column[String]("ingredients")
def * = (id, name, instructions, ingredients) <> ((Recipe.apply _).tupled, Recipe.unapply _)
}
val recipes = TableQuery[Recipes]
val setup = DBIO.seq(
recipes.schema.create,
recipes += Recipe(None, "Chicken with Avocado", "Mix it up", "Chicken, Avocado")
)
def apply() = {
val db = Database.forConfig("h2mem1")
try db.run(setup)
finally db.close
}
}
Or adding .? to the field id, whithout using Option, like we talk in comments, I think that this aproach is better
package org.example
import slick.driver.H2Driver.api._
object SlickStackOverflow extends App {
}
case class Recipe(id: Option[Int], name: String, instructions: String, ingredients: String)
object AddFixtures {
class Recipes(tag: Tag) extends Table[Recipe](tag, "recipe") {
def id = column[Int]("id", O.PrimaryKey, O.AutoInc)
def name = column[String]("name")
def instructions = column[String]("instructions")
def ingredients = column[String]("ingredients")
def * = (id.?, name, instructions, ingredients) <> ((Recipe.apply _).tupled, Recipe.unapply _)
}
val recipes = TableQuery[Recipes]
val setup = DBIO.seq(
recipes.schema.create,
recipes += Recipe(None, "Chicken with Avocado", "Mix it up", "Chicken, Avocado")
)
def apply() = {
val db = Database.forConfig("h2mem1")
try db.run(setup)
finally db.close
}
}
The field1 ~ field2
construct is actually constructing a (field1, field2)
tuple under the hood, so as @anquegi points out, simply changing your *
projection to use a tuple directly will work.
Alternatively, if you want to use ~
to construct the tuple you can get it back by importing TupleMethods
(as ~
was moved out of the normal import scope in Slick 2.0.):
import slick.util.TupleMethods._
See also: Slick 2.0 - update two or more columns
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With