I'm trying to upgrade my project to slick 3.3.0, so I set slick-pg to 0.17.2 but unfortunately in slick-pg 0.17.1 a few deps including json4s were upgraded to latest version. Alas json4s 3.6.x versions have a longstanding bug for case classes with type constructors (json4s/json4s#507). My project also dependes on json4s and thus breaks when upgrading. Is there anyway this dependency hell can be circumvented?
`import com.github.tminglei.slickpg._
trait MyPostgresDriver
extends ExPostgresProfile
with PgArraySupport
with PgDateSupportJoda
with PgEnumSupport
with PgRangeSupport
with PgHStoreSupport
with PgSearchSupport
with PgPostGISSupport {
override val api = new MyAPI {}
trait MyAPI
extends API
with ArrayImplicits
with DateTimeImplicits
with RangeImplicits
with HStoreImplicits
with SearchImplicits
with PostGISImplicits
with SearchAssistants
}
object MyPostgresDriver extends MyPostgresDriver`
Bascially none of the PG types here are found.
org.postgresql.util.PSQLException: ERROR: function st_dwithin(point, bytea, double precision) does not exist
[info] Hint: No function matches the given name and argument types. You might need to add explicit type casts.
[info] Position: 140
def byDistance(point: Point, distance: Double): Future[Seq[House]] =
db run {
houses
.filter(r => r.location.dWithin(point.bind, distance.bind))
.result
.map(t => t)
}
due to: org.postgresql.util.PSQLException: ERROR: column "location" is of type point but expression is of type bytea
[info] Hint: You will need to rewrite or cast the expression.
[info] Position: 157
def distance[P2, R](geom: Rep[P2])(implicit om: o#to[Float, R]) = {
om.column(GeomLibrary.Distance, n, geom.toNode)
}
Rather than repeating myself here, I'll post my https://twitter.com/heyrutvik/status/1182521309991002112 which describes the problem of using
<-> with hyphened string in to_tsquery function.Please take a look, thanks!
drop table if exists orders cascade;
CREATE TABLE orders (
ID serial NOT NULL PRIMARY KEY,
info json NOT NULL
);
INSERT INTO orders (info)
VALUES
('{"interestedIn":[11],"countries":["IT", "UK"]}'),
('{"interestedIn":[12],"countries":["US", "UK"]}'),
('{"interestedIn":[1,2,3],"countries":["UK"]}'),
('{"interestedIn":[14,15,16],"countries":["DE", "UK"]}');
select *
from orders
where info::jsonb -> 'countries' ? 'US'
;
Hi All, we're trying to get upserts working with slick. Trying to use slick-pg to enable this, but we can't find how to access the new methods. We created our own profile that extends ExPostgresProfile, and it looks like it's configured OK in application.conf. We replaced import profile.api._ with import MySlickPostgresProfile.api._
However, our methods that use testtable.insertOrUpdate still seems to use the old version, and .insertOrUpdateAll isn't found. Is there some weird implicit scala thing we need to do?
Option[scala.collection.Seq] which is a type-constructor and not a type, hence the compilation fails
Option[List[String]] - which seems consistent with array columns declaration for slick-pg, as of https://github.com/tminglei/slick-pg#configurable-typemappers - then the generated code comes as expected, but the compiler hangs in some non-terminating loop when trying to compile the project
I am new to slick and i am trying to convert a sql query to slick and need help for the same.
My table is:
final case class TableExp(
id: String,
key: String,
value: String
)
final class TableExp(tag: Tag)(implicit val schema: String)
extends Table[TableExp](tag, Some(schema), "table") {
def id = column[String]("id")
def key = column[String]("key")
def value = column[String]("value")
def * =
(id,
key,
value
).shaped <> (TableExp.tupled, TableExp.unapply)
def getData(id: String) =
{
TableQuery[TableExp].filter(res => res.id === id || res.id === "id1").map(_.key).distinct
}
}And the sql query is:
select distinct key, first_value(value) over (partition by key order by case when t.id=$id then 0 else 1 end) value from table t where t.id in ($id, 'id1')) t
So far i have tried this:
def getData(id: String) =
{
TableQuery[TableExp].filter(res => res.id === id || res.id === "id1").groupBy(_.key).map
{
case (a,b) => Over.partitionBy(b.map(_.key))
}
}
CREATE TABLE mytable(myelements JSONB DEFAULT '[]'::JSONB);
SELECT myelement
FROM mytable,
jsonb_array_elements(mytable.myelements) myelement;
Hi @all I am getting this very wired exception when added the
val slickPgCirciJson = "com.github.tminglei" %% "slick-pg_circe-json" % "0.19.3"[error] java.lang.NoSuchMethodError: com.github.tminglei.slickpg.array.PgArrayJdbcTypes$SimpleArrayJdbcType.to(Lscala/Function1;Lscala/reflect/ClassTag;)Lslick/jdbc/JdbcTypesComponent$DriverJdbcType;and this is how my database class looks like -
trait DatabaseApi extends ExPostgresProfile
with PgArraySupport
with PgDate2Support
with PgRangeSupport
with PgHStoreSupport
with PgCirceJsonSupport
with PgSearchSupport
with PgPostGISExtensions
with PgNetSupport
with PgLTreeSupport {
override val api = PostgresAPI
object PostgresAPI
extends API
with ArrayImplicits
with DateTimeImplicits
with JsonImplicits
with NetImplicits
with LTreeImplicits
with RangeImplicits
with HStoreImplicits
with SearchImplicits
with SearchAssistants
with CirceImplicits {
val coalesce: (Rep[Option[Long]], Rep[Option[Long]]) => Rep[Option[Long]] =
SimpleFunction.binary[Option[Long], Option[Long], Option[Long]](
"coalesce")
implicit val strListTypeMapper: DriverJdbcType[List[String]] =
new SimpleArrayJdbcType[String]("text")
.to(_.toList)
def pgIntervalStr2Interval(intervalStr: String): Interval =
Interval.fromPgInterval(new PGInterval(intervalStr))
}
}
object DatabaseApi extends DatabaseApi {
def pgjson = "jsonb" // jsonb support is in postgres 9.4.0 onward; for 9.3.x use "json"
// Add back capabilities.insertOrUpdate to enable native upsert support; for postgres 9.5+
override protected def computeCapabilities: Set[Capability] =
super.computeCapabilities + JdbcCapabilities.insertOrUpdate
}
:wave: - I ran into an issue with \u0000 characters not being stripped out when using the default PgJsonSupport.
I noticed that the serde-specific artifacts (play-json, circe, etc.) already strip this character out via JsonUtils.clean, would it make sense to add that same stripping logic to PgJsonSupport as well? I'd be happy to open a PR to make that change!