always_join

Usage

explore: explore_name {
  always_join: [
    view_name,
    view_name,
    ...
  ]
}

Definition

always_join forces one or more joins to be included in the SQL that Looker generates, even if the user has not selected a field from that joined view. Multiple joins can be required by using a comma-separated list like [view_name_a, view_name_b, etc].

When Looker generates SQL for a query, it attempts to create the cleanest SQL possible, and will only use the joins that are necessary for the fields a user selects. By using always_join, you can force joins to occur no matter what.

always_join may be valuable when a join is executed with the type parameter, and the join is not a LEFT JOIN. In such a situation the join may be critical to correctly limiting the rows that are returned.

Examples

Make sure that member is always joined to event, even if the user does not choose a field from member. This limits the results to only look at member-generated events:

explore: event {
  always_join: [member]
  join: member {
    sql_on: ${event.member_id} = ${member.id} ;;
    type: inner
  }
}

Make sure that member and payment are always joined to event, even if the user does not choose a field from either of those views. This limits the results to only look at member-generated events where the member has paid already:

explore: event {
  always_join: [member, payment]
  join: member {
    sql_on: ${event.member_id} = ${member.id} ;;
    type: inner
  }
  join: payment {
    sql_on: ${member.payment_id} = ${payment.id} ;;
    type: inner
  }
}

Common challenges

A view must be joined to an Explore before it can be referenced in always_join

To place a view into always_join, make sure that it is joined to the Explore where the always_join is being used. For example, this will not work:

explore: event {
  always_join: [member]
}

Here the member view hasn't been joined to event, so it isn't available for use in always_join.

Things to know

Don't apply business logic in joins if possible

The standard Looker approach to joining is to use a LEFT JOIN whenever possible. In the above examples, we avoid a LEFT JOIN so that business logic can be applied within the join itself. In one of the examples we created an Explore that included only those events that were associated with members:

explore: event {
  always_join: [member]
  join: member {
    sql_on: ${event.member_id} = ${member.id} ;;
    type: inner
  }
}

The preferred way to execute this in Looker is to use a LEFT JOIN to get event data and member data stuck together simply:

explore: event {
  join: member {
    sql_on: ${event.member_id} = ${member.id} ;;
  }
}

Then you could create a dimension that you could set to yes or no, to look only at member events:

dimension: is_member_event {
  type: yesno
  sql: ${member.id} IS NOT NULL ;;
}

This approach gives users the flexibility to look at all events, or only member events. You have not forced users to only look at member events via the join.