What type of enhancement is this?
API improvement
What subsystems and features will be improved?
Command processing, Continuous aggregate
What does the enhancement do?
It took me a little bit to figure out how to do continuous aggregate regression using time as the X variable. Given that it's a time series database, it could make sense to have built-in handling for timestamps as a dependent variable.
i.e. add a function overload for stats_agg(real, timestamp) input types.
The current solution for me was
slope(stats_agg(some_variable::real, extract(epoch from time AT TIME ZONE 'UTC')))
which is a bit verbose and I'm unsure if there are performance implications or if the query planner can de-duplicate them fine.
Implementation challenges
No response