CODEWITHSUNDEEP
rtidyr
Damien Dotta
Damien Dotta

Reputation: 939

How to pivot a tibble?

I'm trying to pivot this tibble with {tidyr} but withouT success...
Can someone help me ?
Many thanks in advance

library(tibble)

mydf <- tibble(
  annee_methode=c("2016m2017","2017m2017"),
  siren_ent=c("A","B"),
  categorie_ent=c("PME","ETI"),
  Prod=c(10,5),
  Intens=c(2,3),
  vaht=c(100,50),
  effeqtp_ent=c(10,10),
  immo_corp=c(20,30)
)

# Using this tibble to determine `Numerateur` and `Denominateur`
calcul_ratios <- tibble(
  ratio = c("Prod","Intens"),
  Numerateur = c("vaht","immo_corp"),
  Denominateur = c("effeqtp_ent","effeqtp_ent")
)

# What I would like :  
# A tibble: 2 x 7
  Libelle Valeur_2016m2017 Numerateur_2016m2017 Denominateur_2016m2017 Valeur_2017m2017 Numerateur_2017m2017 Denominateur_2017m2017
  <chr>              <dbl>                <dbl>                  <dbl>            <dbl>                <dbl>                  <dbl>
1 Prod                  10                  100                     10                5                   50                     10
2 Intens                 2                   20                     10                3                   30                     10

Upvotes: 1

Views: 69

Answers (1)

stefan
stefan

Reputation: 125208

You could do:

library(dplyr)
library(tidyr)

mydf %>% 
  select(-siren_ent, -categorie_ent) %>% 
  pivot_longer(-annee_methode) %>% 
  left_join(pivot_longer(calcul_ratios, everything()), by = c("name" = "value")) %>% 
  mutate(Libelle = rep(c("Prod","Intens"), 6),
         name.y = ifelse(name.y == "ratio", "Valeur", name.y)) %>%
  select(-name) %>% 
  pivot_wider(names_from = c(name.y, annee_methode), values_from = value)
#> # A tibble: 2 x 7
#>   Libelle Valeur_2016m2017 Numerateur_2016m2… Denominateur_201… Valeur_2017m2017
#>   <chr>              <dbl>              <dbl>             <dbl>            <dbl>
#> 1 Prod                  10                100                10                5
#> 2 Intens                 2                 20                10                3
#> # … with 2 more variables: Numerateur_2017m2017 <dbl>,
#> #   Denominateur_2017m2017 <dbl>

Upvotes: 2

Related Questions