Floating-point imprecision can lead to nasty bugs when attempting to
compare epoch boundaries. This allows us to specify a precision when
performing an epoch_intersection operation.
Also fixes a subtle bug when end of an epoch in b matches start of an
epoch in a.
Floating-point imprecision can lead to nasty bugs when attempting to compare epoch boundaries. This allows us to specify a precision when performing an epoch_intersection operation.
Also fixes a subtle bug when end of an epoch in b matches start of an epoch in a.