For example, in pd.date_range, we deprecated some frequency aliases. With pandas 2.3:

>>> pd.date_range("2012-01-01", periods=3, freq='H') FutureWarning: 'H' is deprecated and will be removed in a future version, please use 'h' instead. DatetimeIndex(['2012-01-01 00:00:00', '2012-01-01 01:00:00', '2012-01-01 02:00:00'], dtype='datetime64[ns]', freq='h') 

and now with main (pandas 3.0) the deprecation is enforeced and we raise an error:

>>> pd.date_range("2012-01-01", periods=3, freq='H') ... ValueError: Invalid frequency: H, failed to parse with error message: ValueError("Invalid frequency: H, failed to parse with error message: KeyError('H')") 

However, I think that is not the best error message ..
First, it's already a bit confusing that it contains the message twice.
But secondly, I think it would also be nice to still keep a mapping of those aliases around, so that when parsing fails and the input is an old alias, we can provide a more informative error message (something like "invalid frequency 'H'. Did you mean 'h'?")