Instruction
Great! Of course, you can partition rows by multiple columns. Take a look:
SELECT route_id, ticket.id, ticket.price, SUM(price) OVER (PARTITION BY route_id, date) FROM ticket JOIN journey ON ticket.journey_id = journey.id;
We wanted to show each ticket with the sum of all tickets on the particular route on the particular date. Neither of the tables would suffice on its own, so we had to join them together to get all the columns.
Exercise
Show the id of each journey, the date on which it took place, the model of the train that was used, the max_speed of that train and the highest max_speed from all the trains that ever went on the same route on the same day.
Stuck? Here's a hint!
Use
MAX(max_speed) OVER(PARTITION BY route_id, date)



