UT1 incorrect during leap second when using EOP
When using EOP corrections for UT1-UTC I would expect that UT1-TAI is smooth, but instead there are two step discontinuities at the start and end of a leap second, as seen in the plot below. This creates discontinuities in the inertial to fixed frame transformation around a leap second. The cause seems to be that UTCScale.offsetFromTai()
increments at the start of the leap second, while EOPHistory.getUT1MinusUTC()
considers the discontinuity to occur at the end of the leap second. This leaves 1 second where UT1-UTC is incorrect and hence Earth rotation is incorrect.