> If you mean that your input will always include an explicit zone > specification, then this doesn't affect you. What I was thinking > about was that > > select generate_series('2023-10-04 13:30', ...)
> is going to mean different things depending on the zone setting > that prevails when that constant is parsed.
Gotcha--I thought you meant the timezone argument to the function ($4). I can make sure that all the datetime arguments to the function are "with timezone", so there should be no ambiguity. But I guess if I didn't, the timezone given by $4 would be the one parsing the naive timestamps, which I think would also be ok from my perspective.
