Currently, condense_segments returns the original DataFrame passed in, but overwrites the start_dt_col and end_dt_col values. Seems like the function should have 2 options, similar to extend_segments:
Return the original DataFrame, with 2 additional columns for start/end dates
Return a new DataFrame, only containing the group_col columns and the new start/end date columns
Right now, condense_segments is somewhere in-between. If a user wants option 2, they have to re-select the grouping columns and date columns then call distinct(). If a user wants option 1, they have to join back to the original DataFrame.
Adding a retain_shape parameter would simplify user code and enable both options.
Currently,
condense_segmentsreturns the original DataFrame passed in, but overwrites thestart_dt_colandend_dt_colvalues. Seems like the function should have 2 options, similar toextend_segments:group_colcolumns and the new start/end date columnsRight now,
condense_segmentsis somewhere in-between. If a user wants option 2, they have to re-select the grouping columns and date columns then calldistinct(). If a user wants option 1, they have to join back to the original DataFrame.Adding a
retain_shapeparameter would simplify user code and enable both options.