Maybe something like this? https://go.dev/play/p/umMkwm4nBiE

If you're able to link up any Go code in the wild (eg from the Go source or other open source projects) that are doing this kind of calculation currently, or any packages out there that implement it, that would probably help further guide discussion and possible implementation.

Thanks @danp at a glance that looks along the right lines.

Unfortunately for all their boasts about being postgres compatible, Cockroach have not implemented date_bin() otherwise that would have been the obvious place to check !

I'm a bit busy with $work at the moment, but I'll aim to find some time to both revisit your example and also see if I can find any open source stuff lurking around.

The proposal is

// Bin rounds the time down to the nearest multiple of d offset from the start time.
func (t Time) Bin(d Duration, start time.Time) time.Time

This is a reasonably simple function. The question is how often it comes up.

Adding to minutes. The question remains how commonly needed this is.

This proposal has been added to the active column of the proposals project
and will now be reviewed at the weekly proposal review meetings.
— rsc for the proposal review group

As noted earlier, this function is pretty short and easy to write outside the standard library:

func Bin(t time.Time, d time.Duration, origin time.Time) time.Time {
	r := t.Sub(origin) % d
	if r < 0 {
		r += d
	}
	return t.Add(-r)
}

This one does have the downside of not working if t.Sub(origin) overflows or underflows, which happens if the origin time is more than 584 years away from the current time. For any actual use of Bin that's not going to happen, so this code is fine in a third-party library. (If it was in package time we'd have to deal with that case.)

I've asked twice for how often this comes up with little response, so it sounds like maybe we should leave this for implementation outside the standard library.

Based on the discussion above, this proposal seems like a likely decline.
— rsc for the proposal review group

No change in consensus, so declined.
— rsc for the proposal review group