Idea: implement set_var_labels_with

[ABohynDOE]

Firstly, I really enjoy this package ! I mainly use the set_var_labels() function to define variable labels to obtain nice summary tables with gtsummary.

However, I often think that it would be useful to have a function that sets variable labels based on the column names and that can be applied in the same way as rename_with() works.

For example, if I have a data frame with columns quest_1 to quest_10, and location_A to location_Z, I wish there was a function I could use to do:

labelled_df <- df %>%
  set_var_labels_with(
    .fn = ~paste0("Question ", str_extract(.x, "\\d+$")),
    .cols = starts_with("quest_")
  ) %>% 
  set_var_labels_with(
    .fn = ~paste0("Location ", str_extract(.x, ".$")),
    .cols = starts_with("location_")
  )

Of course this is maybe a silly example, but I think that it could allow labelling variables within a pipe workflow, without having to manually generate labels when there are many columns.

[larmarange]

There is already an update_variable_labels_with() function provided by labelled. With #164, variable names are now accessible using names().

library(labelled)

iris |> 
  update_variable_labels_with(~ toupper(names(.x))) |> 
  look_for()
#>  pos variable     label        col_type missing values    
#>  1   Sepal.Length SEPAL.LENGTH dbl      0                 
#>  2   Sepal.Width  SEPAL.WIDTH  dbl      0                 
#>  3   Petal.Length PETAL.LENGTH dbl      0                 
#>  4   Petal.Width  PETAL.WIDTH  dbl      0                 
#>  5   Species      SPECIES      fct      0       setosa    
#>                                                 versicolor
#>                                                 virginica

^{Created on 2024-10-03 with reprex v2.1.1}

[ABohynDOE]

OK thanks a lot, I didn't find this function earlier !