cutData can overwrite earlier created columns with unexpected data

[mooibroekd]

Bug description

Today I tracked down a bug where the 'year' column in my data was a bit strange. It turns out that cutData overwrites the data when a particular order is used to create columns.

Steps to reproduce

packageVersion("openair")
#> [1] '2.18.2'

# create some date data for illustration
test <- data.frame(date = seq.Date(lubridate::date("2022-01-01"), 
                                   lubridate::date("2022-12-31"), 
                                   by = 1))

# cut the data into years
test <- openair::cutData(test, type = "year")

# check if we have any dates that are associated with 2023
test |>
  dplyr::filter(year == 2023)
#> [1] date year
#> <0 rows> (or 0-length row.names)

# now cut the data to yearseason
test <- openair::cutData(test, type = "yearseason")

# perform the test again:
test |>
  dplyr::filter(year == 2023)
#>          date year       season month          yearseason
#> 1  2022-12-01 2023 winter (DJF)    12 winter (DJF) - 2023
#> 2  2022-12-02 2023 winter (DJF)    12 winter (DJF) - 2023
#> 3  2022-12-03 2023 winter (DJF)    12 winter (DJF) - 2023
#> 4  2022-12-04 2023 winter (DJF)    12 winter (DJF) - 2023
#> 5  2022-12-05 2023 winter (DJF)    12 winter (DJF) - 2023
#> 6  2022-12-06 2023 winter (DJF)    12 winter (DJF) - 2023
#> 7  2022-12-07 2023 winter (DJF)    12 winter (DJF) - 2023
#> 8  2022-12-08 2023 winter (DJF)    12 winter (DJF) - 2023
#> 9  2022-12-09 2023 winter (DJF)    12 winter (DJF) - 2023
#> 10 2022-12-10 2023 winter (DJF)    12 winter (DJF) - 2023
#> 11 2022-12-11 2023 winter (DJF)    12 winter (DJF) - 2023
#> 12 2022-12-12 2023 winter (DJF)    12 winter (DJF) - 2023
#> 13 2022-12-13 2023 winter (DJF)    12 winter (DJF) - 2023
#> 14 2022-12-14 2023 winter (DJF)    12 winter (DJF) - 2023
#> 15 2022-12-15 2023 winter (DJF)    12 winter (DJF) - 2023
#> 16 2022-12-16 2023 winter (DJF)    12 winter (DJF) - 2023
#> 17 2022-12-17 2023 winter (DJF)    12 winter (DJF) - 2023
#> 18 2022-12-18 2023 winter (DJF)    12 winter (DJF) - 2023
#> 19 2022-12-19 2023 winter (DJF)    12 winter (DJF) - 2023
#> 20 2022-12-20 2023 winter (DJF)    12 winter (DJF) - 2023
#> 21 2022-12-21 2023 winter (DJF)    12 winter (DJF) - 2023
#> 22 2022-12-22 2023 winter (DJF)    12 winter (DJF) - 2023
#> 23 2022-12-23 2023 winter (DJF)    12 winter (DJF) - 2023
#> 24 2022-12-24 2023 winter (DJF)    12 winter (DJF) - 2023
#> 25 2022-12-25 2023 winter (DJF)    12 winter (DJF) - 2023
#> 26 2022-12-26 2023 winter (DJF)    12 winter (DJF) - 2023
#> 27 2022-12-27 2023 winter (DJF)    12 winter (DJF) - 2023
#> 28 2022-12-28 2023 winter (DJF)    12 winter (DJF) - 2023
#> 29 2022-12-29 2023 winter (DJF)    12 winter (DJF) - 2023
#> 30 2022-12-30 2023 winter (DJF)    12 winter (DJF) - 2023
#> 31 2022-12-31 2023 winter (DJF)    12 winter (DJF) - 2023

# reason why this is happening is because the winter is defined from december 
# yyyy to march yyyy+1.
# I would suspect cutData to only return the yearseason, not the year and season
# columns (or overwrite them if they exists.)

^{Created on 2024-12-22 with reprex v2.1.1}

Openair version

No response