Across attempts

Must have latest version of dplyr.

library(tidyverse)

Using if_any to mutate multiple columns

I want to classify rows in a new column based on any value in a list being present in any column in a list. But I want to do this using across() or if_any().

This is what I’m looking for, listing each column in the case_when(). I want to replace multiple lines with one if_any(my_color_cols).

color_values <- c("blue", "brown")
color_cols <- c("hair_color", "skin_color", "eye_color")

starwars %>% 
  select(name, ends_with("color")) %>% 
  mutate(
    blue_brown = case_when(
      hair_color %in% color_values ~ TRUE,
      skin_color %in% color_values ~ TRUE,
      eye_color %in% color_values ~ TRUE,
      TRUE ~ FALSE
    )
  )

This is the if_any() rewrite of the above code.

starwars %>% 
  select(name, ends_with("color")) %>% 
  mutate(
    blue_brown = if_any(color_cols, ~ case_when(
      .x %in% color_values ~ TRUE,
      TRUE ~ FALSE
    ))
  )

## Note: Using an external vector in selections is ambiguous.
## ℹ Use `all_of(color_cols)` instead of `color_cols` to silence this message.
## ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
## This message is displayed once per session.

---

Using smaller data set

Import

data <- read_csv("data/data.csv")

## 
## ── Column specification ────────────────────────────────────────────────────────────────────────────────
## cols(
##   id = col_double(),
##   color_a = col_character(),
##   color_b = col_character(),
##   color_c = col_character()
## )

data

Variables

my_color_cols <- c("color_a", "color_b", "color_c")

Mutate with across

case_when method for mutate

This works. It finds TRUE when either blue or brown is an any of the three columns.

data %>% 
  mutate(
    blue_red = case_when(
    color_a %in% color_values ~ TRUE,
    color_b %in% color_values ~ TRUE,
    color_c %in% color_values ~ TRUE,
    TRUE                     ~  FALSE
    )
  )

if_any version of same task

This uses if_any() to look across my_color_cols to find any value in color_valus and mark as TRUE in a new column.

data %>% 
  mutate(
    blue_red = if_any(all_of(my_color_cols), ~ case_when(
    .x %in% color_values ~ TRUE,
    TRUE ~  FALSE
    ))
  )

It seems strange to use both if_any(all_of(my_color_rows)), but without it I got this warning:

Note: Using an external vector in selections is ambiguous.
ℹ Use `all_of(my_color_cols)` instead of `my_color_cols` to silence this message.
ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.

So I’ll try this one more time to see if I get the same error, but using a select-type value?

data %>% 
  mutate(
    blue_red = if_any(starts_with("color"), ~ case_when(
    .x %in% color_values ~ TRUE,
    TRUE ~  FALSE
    ))
  )

Yes … perhaps the key note there is “external vector” since originally I was calling my column names from a vector created outside the data frame.