The Python version simply forces the timestamp forward if needed to preserve ordering:

  if timestamp_ms < _last_timestamp_v7:
    timestamp_ms = _last_timestamp_v7 + 1

But if you did want "monotonically increasing IDs", as the article suggests, that is what you want.

So you can't generate more than 1000 UUIDs per second without skewing timestamp portion? That's not a lot. Bulk imports often require a lot more inserts.

This is perfect if you only have one machine and the process never restarts. But then if you only have one machine you don’t realistically need uuid identifiers.