CODEWITHSUNDEEP
rdataframedplyr
Homer Jay Simpson
Homer Jay Simpson

Reputation: 1280

How can I match the values of a column according to another data frame in R and print a message using dplyr?

I have a dataset that looks like this :

id1 var1
A chair
B table
C glass
D phone
E pistol

and a second one that contains the licenses of each id (but it could contain different ids)

id2 var2
G mobile
H pistol
I pistol
E phone
D phone

I want to check if the ids in the first dataframe are licensed to have what they declare. For example the id D is licensed to have a phone but E is not licensed to have pistol because it's licensed to have phone. So there are three conditions here and ideally the final data frame must look like this:

id1 var1 license
A chair not_needed
B table not_needed
C glass not_needed
D phone ok_checked
E pistol danger

How can I make this cross check and print these messages according to the logical condition in R using dplyr?

library(tidyverse)
id1 =c("A","B","C","D","E")
var1 = c("chair","table","glass","phone","pistol")
data1 = tibble(id1,var1);data1
id2 = c("G","H","I","E","D")
var2 = c("mobile","pistol","pistol","phone","phone")
data2 = tibble(id2,var2);data2

Upvotes: 0

Views: 832

Answers (2)

Asitav Sen
Asitav Sen

Reputation: 76

library(dplyr)
data1 |> 
  full_join(data2, by=c("id1"="id2")) |> 
  rename(declared=var1, actual=var2) |> 
  mutate(license=ifelse(is.na(declared),"Not declared",
                        ifelse(declared %in% c("chair","table","glass"),"Not needed",
                               ifelse(declared==actual,"OK","Danger"))))

I have use full_join. You can use let join if you don't need all the ids. I have also assumed that chair, table and gas do not need license. You can add or remove items for your need. Finally, you can remove the columns that you don't need. It is also possible to use case_when() function instead of ifelse() statements to achieve the same result.

  id1   declared actual license     
  <chr> <chr>    <chr>  <chr>       
1 A     chair    NA     Not needed  
2 B     table    NA     Not needed  
3 C     glass    NA     Not needed  
4 D     phone    phone  OK          
5 E     pistol   phone  Danger      
6 G     NA       mobile Not declared
7 H     NA       pistol Not declared
8 I     NA       pistol Not declared

Upvotes: 1

benson23
benson23

Reputation: 19097

You can first left_join the two dataset, then use a case_when statement to assign terms to the condition.

library(tidyverse)

left_join(data1, data2, by = c("id1" = "id2")) %>% 
  mutate(var2 = case_when(is.na(var2) ~ "not_needed", 
                          var1 == var2 ~ "ok_checked", 
                          var1 != var2 ~ "danger", 
                          TRUE ~ NA_character_)) %>% 
  rename("license" = "var2")

# A tibble: 5 × 3
  id1   var1   license   
  <chr> <chr>  <chr>     
1 A     chair  not_needed
2 B     table  not_needed
3 C     glass  not_needed
4 D     phone  ok_checked
5 E     pistol danger

Upvotes: 1

Related Questions