creating a column function with custom parameters-CodePudding

I've got a question here:

Create a dummy dataframe ' df_test' of 10 rows comprising of column (1) ' INT1' with sample values between the range 1 and 10

Create a function ' check_even' that accepts two arguments

(1) ' df' - a dataframe

(2) ' col_test' - name of column to test if each row value in the indicated column is even or odd.

(3) ' col_multiply' - name of the column that stores the result of 'col_test' multiply by 2 if the 'col_test' is even and multiply by 3 if the 'col_test' is odd

(4) the function ends by returning the entire df with the results

Test your function by running this code ' check_even(df_test, 'INT1', 'res')'

Here are my codes:

df_test = data.frame(INT1 = (sample(x = c(1:10),size = 10, replace = F)))
df_test

check_even = function(df, col_test, col_multiply) {
    df$res = ifelse(df[col_test,] %% 2 == 0, df[,col_multiply] * 2, df[,col_multiply] * 3)
    return(df)} #wrong

#run code
check_even(df_test, 'INT1','res')

In the code above, I get:

I'd also tried another way to do this using dplyr:

df_test = data.frame(INT1 = (sample(x = c(1:10),size = 10, replace = F)))
df_test

library(dplyr)

    check_even = function(df, col_test, col_multiply){
      df %>%
        mutate(res = ifelse({{col_test}} %% 2 == 0, {{col_multiply}} * 2, {{col_multiply}} * 3))
    }  
    
    check_even(df_test, 'INT1', 'res')

However, I'm still getting the error :

Error in mutate(., res = ifelse({ : Caused by error in "INT1" %% 2: ! non-numeric argument to binary operator

How can I solve this question? Thank you in advance.

CodePudding user response：

Base R

Here is the base R function corrected.

df_test = data.frame(INT1 = (sample(x = c(1:10),size = 10, replace = FALSE)))
#df_test

check_even = function(df, col_test, col_multiply) {
  df[[col_multiply]] = ifelse(df[[col_test]] %% 2 == 0, df[[col_test]] * 2, df[[col_test]] * 3)
  return(df)
}

#run code
check_even(df_test, 'INT1','res')
#>    INT1 res
#> 1     4   8
#> 2     5  15
#> 3     1   3
#> 4     8  16
#> 5     2   4
#> 6     9  27
#> 7     3   9
#> 8     7  21
#> 9    10  20
#> 10    6  12

^{Created on 2022-06-04 by the reprex package (v2.0.1)}

`dplyr` solution

library(dplyr)

check_even = function(df, col_test, col_multiply){
  df %>%
    mutate({{col_multiply}} := ifelse({{col_test}} %% 2 == 0, {{col_test}} * 2, {{col_test}} * 3))
}  

check_even(df_test, INT1, res)

CodePudding user response：

I think you need this, if I am not wrong you should change your logic:

Instead of passing the result column as an argument you could calculate it withn the function body:

library(dplyr)

check_even = function(df, col_test){
  
  df %>%
    mutate(res = ifelse({{col_test}} %% 2 == 0, {{col_test}} * 2, {{col_test}} * 3))
}  

check_even(df_test, INT1)

   INT1 res
1    10  20
2     3   9
3     6  12
4     4   8
5     7  21
6     9  27
7     5  15
8     2   4
9     1   3
10    8  16

Base R

dplyr solution

`dplyr` solution