R regex extract standalone characters

Question

I'm trying to extract letter R or O that stands alone from multiple columns. By standalone, I mean R or O (i) separated by space or (ii) that is the only value in a cell. Here's a reproducible example. Suppose I want to extract standalone R or O from column X1 and X2 .

df <- data.frame(X1 = c( "EHO", "X 1 R","R"),
                 X2 = c( "Y R E", "X A 1", "AER"), 
                 X3 = NA)

Here's desired outcome.

data.frame(X1 = c("", "R", "R"),
           X2 = c("R", "", ""))

Here's what I've tried so far. The first approach is problematic because R from AER and O from EHO is extracted (also R from "Y R E" is not extracted).

require(stringr)
sapply(df[,1:2], function(x) ifelse( df$X3 %in% NA, str_extract(x, "\\s?[O|R]$"), X3))

So I've tried this, which solves above problem, but now it fails to extract R from df[3,1] .

sapply(df[,1:2], function(x) ifelse( df$X3 %in% NA, str_extract(x, "(?![A-Z]+?)\\s?[O|R]$?"), X3))

How should I fix the pattern to get this?

Answer 1

You can use word boundaries:

sapply(df, stringr::str_extract, '\\b[RO]\\b')

#     X1  X2  X3
#[1,] NA  "R" NA
#[2,] "R" NA  NA
#[3,] "R" NA  NA

However, note that str_extract will extract only one of "R" or "O" whichever comes first.

stringr::str_extract('EH R O', '\\b[RO]\\b')
#[1] "R"

If you want to extract both of them you might need to use str_extract_all .