Should epoch parameter of eras also have an era?

[joeroe]

Currently era() assumes the epoch parameter is a year in the Common Era, but we could make this itself a yr object (defaulting to CE).

[joeroe]

If this is implemented we would have to change the hard-coded length of a year in yr_transform() to calls to era_year_days(era_unit(era(era_epoch())). But I'm no longer sure about this. The benefit seems marginal, it will probably introduce some recursion issues, and will definitely introduce more complexity for users when defining eras.