error when unique! a empty dataframe

[1991jhf]

I get inexact error trunc(Int64,Inf) when unique! an empty dataframes.

It track down to

function row_group_slots!(
...
# if there are few rows per group limit the number of threads used
nt = clamp(round(Int, (lg / 4) / ngroups - 2), 1, nt)
...

when ngroups=0.
However i cannot reproduce this in a minimal example.
Is it possible to short-circuit the unique! when the dataframe is empty?

Thank you

[bkamins]

fixed in #3393