Count of dplyr is not working properly inside a function-CodePudding

I am trying to print aggregate tables and plots for different columns but when i pass a column name it is saying 'object is not found'. Here is a simple example:

df <- data.frame(
  "age" = c(1,2,3),
  "name" = c("a", "b", "c")
)
l <- c("age", "name")

#function to loop through columns
func <- function(df, parameter){
  return(
    df %>% 
      count(parameter)
  )
}

When I call the above func like this

func(df, "age")

I am getting this error:

Error in `group_by()`:
! Must group by variables found in `.data`.
✖ Column `parameter` is not found.

What could be the reason? Is it not allowed to pass column name as a string? If not, then how should I proceed? thanks in advance.

CodePudding user response：

You have a typo in your code. paramert is supposed to be parameter.

So change count(paramert) to count(parameter)

CodePudding user response：

You need to turn column name strings into symbols and inject them inside a data-masked expression with !!!. For more details, see the symbolise-and-inject pattern of metaprogramming in R.

func1 <- function(df, parameter){
  return(
    df %>% 
      count(!!!syms(parameter))
  )
}

func1(df, "age")
#   age n
# 1   1 1
# 2   2 1
# 3   3 1

func1(df, c("age", "name"))
#   age name n
# 1   1    a 1
# 2   2    b 1
# 3   3    c 1