Are these differences expected?

[maurolepore]

The intended release of r2dii.data 0.1.5 exposed problems with reverse dependencies: r2dii.match and r2dii.analysis.
The problems in r2dii.match seems related to changes in the column borderline

subset(
  r2dii.data::data_dictionary, 
  subset = column == "borderline", 
  select = c("dataset", "column")
)
#>                   dataset     column
#> 14     cnb_classification borderline
#> 30    gics_classification borderline
#> 38    isic_classification borderline
#> 63    nace_classification borderline
#> 68   naics_classification borderline
#> 91 sector_classifications borderline
#> 95     sic_classification borderline

^{Created on 2020-11-19 by the reprex package (v0.3.0)}

Here I see many differences and I need help to decide if updating the regression references is the right thing to do.

[maurolepore]

@jdhoffa please comment.

[jdhoffa]

The cause for these differences is not immediately clear to me, and somewhat unexpected. I will explore and open a PR if I figure anything out