Read multiple values as character vectors from an external file

March 17, 2024

I am currently reading data from a .csv that looks like this:

param_file <- tribble(
  ~variable, ~value,
  "year", "2023",
  "version", "Saint_XL, Sinner_XY",
  "metric", "ATE, OFCE"
)

I then need to load the variable and its value as character vectors so I am doing this, with dplyr:

year <- param_file %>% filter(variable == "year") %>% pull(value)
version <- param_file %>% filter(variable == "version") %>% pull(value)
metric <- param_file %>% filter(variable == "metric") %>% pull(value)

This works for single values, i.e.
year = "2023",

but the variables with multiple values are being loaded as one value i.e.
"Saint_XL, Sinner_XY"
and
"ATE, OFCE"

Here is the desired output:

> year
[1] "2023"
> version
[1] "Saint_XL"  "Sinner_XY"
> metric
[1] "ATE"  "OFCE"

i.e. character vectors which would otherwise appear in the Environment Values as:
chr[1:2] "Saint_XL" "Sinner_XY"

A dplyr solution would really be appreciated. (Or any useful suggestions around loading variables and assigning values)

>Solution :

With separate_rows():

library(tidyverse)

param_file <- tribble(
  ~variable, ~value,
  "year", "2023",
  "version", "Saint_XL, Sinner_XY",
  "metric", "ATE, OFCE"
)

new_file <- param_file |> 
  separate_rows(value, sep = ",") |> 
  mutate(value = str_squish(value))

year <- new_file %>% filter(variable == "year") %>% pull(value)
version <- new_file %>% filter(variable == "version") %>% pull(value)
metric <- new_file %>% filter(variable == "metric") %>% pull(value)

year
#> [1] "2023"
version
#> [1] "Saint_XL"  "Sinner_XY"
metric
#> [1] "ATE"  "OFCE"

^{Created on 2024-03-17 with reprex v2.1.0}