Query that will benefit from grouping sets

active_contracts.sql

with
  events as
    (select
        modelid as ctr,
        ctime,
        (patch->>'isActive')::bool as isActive
      from "Event"
      where modelname = 'Contract'
        and type = 'Update'
        and (patch->>'isActive')::bool is not null
    ),
  extra_events as
    (select distinct on (ctr)
        ctr,
        current_timestamp,
        not isActive
      from events
      order by ctr, ctime desc
    ),
  active_periods as
    (select distinct
        ctr,
        coalesce(
          lag(e.ctime) over (partition by ctr order by e.ctime), -- max(ctime, from)
          c.ctime) :: date as active_from,
        e.ctime :: date as active_till
      from (select * from events union select * from extra_events) e
        join "Contract" c on (c.id = e.ctr)
      where not e.isActive
    ),
  filtered_periods as
    (select
        ctr,
        daterange(active_from, active_till) *  '[2015-01-01,2016-01-01]' as active_range
      from active_periods
      where
        daterange(active_from, active_till) && '[2015-01-01,2016-01-01]'
    )
  select *
    from
      (select
          p.ctr,
          c.vin,
          make.label,
          model.label,
          c.plateNum,
          null as active_from,
          null as active_till,
          p.total
        from (
            select ctr, sum(upper(active_range) - lower(active_range)) as total
              from filtered_periods
              group by ctr) p
          join "Contract" c on (p.ctr = c.id)
          left join "CarMake" make on (c.make = make.id)
          left join "CarModel" model on (c.model = model.id)
      ) x
      union
      (select
          ctr, null, null, null, null,
          lower(active_range), upper(active_range),
          upper(active_range) - lower(active_range)
        from filtered_periods
      )
    order by ctr, active_from nulls first;