Skip to contents

Read the per-column type-conversion report

Source: R/type_report.R
type_conversion_report.Rd

Returns the record r_data_types attaches to its result: one row per column, naming the rule that fired and where the levels came from.

Usage

type_conversion_report(x)

Arguments

x

An object returned by r_data_types.

Value

A data frame with columns variable, storage_in, rule, level_source, n_levels and storage_out, one row per column of the converted dataset, in column order.

Details

A column that became a factor because it was declared one, and a column that became a factor because it happened to have seven distinct values, are otherwise the same object. This is how a caller tells them apart.

The columns are:

  • variable – the column name.

  • storage_in – its class as received, before conversion.

  • rule – one of "value_labels", "binary_logical", "binary_factor", "character_factor", "n_distinct_factor", "unchanged" or "skipped". "unchanged" means every rule was tested and none fired; "skipped" means the column was named in skip_vars and no rule was tested.

  • level_source – "value labels" where the levels came from the column's own value labels, "inference" where they came from counting distinct values, and NA where no levels were made.

  • n_levels – levels produced, or NA for a non-factor.

  • storage_out – the class of the returned column.

The report is an attribute of the returned object. Operations that rebuild a data frame – most dplyr verbs among them – drop it. Read it directly from the r_data_types() result.

See also

r_data_types

Examples

converted <- r_data_types(datasets::mtcars, use_value_labels = FALSE)
type_conversion_report(converted)
#>    variable storage_in              rule level_source n_levels storage_out
#> 1       mpg    numeric         unchanged         <NA>       NA     numeric
#> 2       cyl    numeric n_distinct_factor    inference        3      factor
#> 3      disp    numeric         unchanged         <NA>       NA     numeric
#> 4        hp    numeric         unchanged         <NA>       NA     numeric
#> 5      drat    numeric         unchanged         <NA>       NA     numeric
#> 6        wt    numeric         unchanged         <NA>       NA     numeric
#> 7      qsec    numeric         unchanged         <NA>       NA     numeric
#> 8        vs    numeric    binary_logical         <NA>       NA     logical
#> 9        am    numeric    binary_logical         <NA>       NA     logical
#> 10     gear    numeric n_distinct_factor    inference        3      factor
#> 11     carb    numeric n_distinct_factor    inference        6      factor