DataPrism

A new FRM with focus on Higher Kinded Data

DataPrism is an SQL query construction library built to take full advantage of the power of higher kinded data. DataPrism builds on perspective and the tools it provides for programming with higher kinded data.

The power of higher kinded data makes DataPrism more flexible than other Scala SQL libraries. Want to sometimes leave out a column? You can do that. Want to return a List from a query? Sure thing.

DataPrism works with both Java's JDBC and Skunk.

Add DataPrism to your project by adding these statements to your build.sbt file.

// For JDBC
libraryDependencies += "net.katsstuff" %% "dataprism-jdbc" % "{{versions.dataprism}}"

// For Skunk
libraryDependencies += "net.katsstuff" %% "dataprism-skunk" % "{{versions.dataprism}}"

DataPrism is currently early in development, but feel free to try it out and report bugs and errors.

Showcase

See the docs for a full explanation and walkthrough on how to use DataPrism.

case class HomeK[F[_]](
  owner: F[UUID],
  name: F[String],
  createdAt: F[Instant],
  updatedAt: F[Instant],
  x: F[Double],
  y: F[Double],
  z: F[Double],
  yaw: F[Float],
  pitch: F[Float],
  worldUuid: F[UUID]
)

object HomeK {

  import dataprism.jdbc.sql.JdbcType
  import dataprism.jdbc.sql.PostgresJdbcTypes.*

  val table: Table[HomeK, JdbcType] = Table(
    "homes",
    HomeK(
      owner = Column("owner", uuid),
      name = Column("name", text),
      createdAt = Column("created_at", javaTime.timestamptz),
      updatedAt = Column("updated_at", javaTime.timestamptz),
      x = Column("x", doublePrecision),
      y = Column("y", doublePrecision),
      z = Column("z", doublePrecision),
      yaw = Column("yaw", real),
      pitch = Column("pitch", real),
      worldUuid = Column("world_uuid", uuid)
    )
  )

  given KMacros.ApplyTraverseKC[HomeK] = KMacros.deriveApplyTraverseKC[HomeK]
}


Query.from(HomeK.table).map(homes => (homes.owner, homes.name))

Query
  .from(HomeK.table)
  .groupBy(homes => (homes.owner, homes.name))((grouped, homes) =>
    (
      grouped._1,
      grouped._2,
      homes.x.arrayAgg,
      homes.y.arrayAgg,
      homes.z.arrayAgg
    )
  )