px_data(labels…) with both code and labels

[larpSTATGL]

Maybe — but this really doesn't belong in this issue. Could you open a new one and close this again? It’s hard to keep track when unrelated things are added to existing issues.

In the new issue, please also describe how you expect the dataframe with labels and codes to look. For example, should each variable have two columns, and what should those columns be called?

With codes and labels one could figure that M, mænd in Danish is M, males and so on.

Can it work with col names is original varnames and labels are named varnames_lbl ?

[johan-ejstrud]

With codes and labels one could figure that M, mænd in Danish is M, males and so on.

Yes that is a smart use case, but isn't this easier to get with px_values()?

For example, if I what to see the codes and labels in all languages for gender I do:

library(pxmake)
library(tidyverse)

x <- px('https://bank.stat.gl:443/sq/a4f0e436-8906-4ba3-a70e-d61527844094')

px_values(x) %>% 
  filter(`variable-code` == "gender")

With a little pivot I can even get it in the format below, which is nice because each code is only listed in one row.

px_values(x) %>% 
  filter(`variable-code` == "gender") %>% 
  pivot_wider(values_from = value, names_from = language)

# A tibble: 3 × 5
  `variable-code` code  en    da      kl        
  <chr>           <chr> <chr> <chr>   <chr>     
1 gender          K     Women Kvinder Arnat     
2 gender          M     Men   Mænd    Angutit   
3 gender          T     Total I alt   Katillugit

Is this what you are looking for, or what do you think? 😊

[larpSTATGL]

Vi lader den vente til måske