3

I have a unnest_tokens function that works in the code, but once I put it into a function I cannot get it to work. I don't understand why this happens when I put it inside a function.

data:

id          words

1           why is this function not working
2           more text
3           help me
4           thank you
5           in advance
6           xx xx

The data is checked on stringsAsFactors == FALSE and if it's a Vector.

is.vector(data$words)
[1] TRUE
is.vector(data$id)
[1] TRUE
typeof(data$words)
[1] "character"

Here is the code outside of the function which gives the correct output:

df <- x %>% 
  unnest_tokens(word, words)%>%
  group_by(id)

1 why
1 is
1 this
1 function
1 not
1 working
2 more
2 text
3 help
3 me
4 thank
4 you
5 in
5 advance
6 xx
6 xx

Once I put the code in a function I get an error.

tidy_x <- unnestDF(data, "words", "id")

unnestDF <- function(df, col, groupbyCol) {
  x <- df %>%
    unnest_tokens(word, df[col])%>%
    group_by(df[groupbyCol])
  return(x)
}

Error in check_input(x) : Input must be a character vector of any length or a list of character vectors, each of which has a length of 1.

Thank you in advance.

Dennis Loos
  • 113
  • 2
  • 9

1 Answers1

4

As we are using quoted arguments, one option would be to convert to symbol and then evaluate (!!) within unnest_tokens and instead of group_by use group_by_at which can take strings

unnestDF <- function(df, col, groupbyCol) {
  df %>%
    unnest_tokens(word, !! rlang::sym(col))%>%
    group_by_at(groupbyCol)

   }


unnestDF(data, "words", "id")
# A tibble: 16 x 2
# Groups:   id [6]
#      id word    
# * <int> <chr>   
# 1     1 why     
# 2     1 is      
# 3     1 this    
# 4     1 function
# 5     1 not     
# 6     1 working 
# 7     2 more    
# 8     2 text    
# 9     3 help    
#10     3 me      
#11     4 thank   
#12     4 you     
#13     5 in      
#14     5 advance 
#15     6 xx      
#16     6 xx      
akrun
  • 874,273
  • 37
  • 540
  • 662