Check if a value in a R dataframe adheres to a column values/any combination of the values in a list-CodePudding

I have a really large dataframe and I need to check that the values for a certain column adhere to an item in a list. This can be any item from the list or a combination of items with a comma separating them.

In the example below, I only want the last value (‘no colour’) to throw a fail as it doesn’t appear in the list called Type.

Type <- list(c('blue','green','black','red'))

Data <-data.frame(colour=c("blue","blue,green", 'blue,black,red', 'black,red', 'no colour'))

Thanks

CodePudding user response：

We may paste the elements in the list and filter

library(stringr)
library(dplyr)
Data %>% 
  filter(str_detect(colour, str_c(Type[[1]], collapse = "|")))

-output

           colour
1           blue
2     blue,green
3 blue,black,red
4      black,red

CodePudding user response：

Using strsplit.

sapply(strsplit(Data$colour, ','), \(x) all(x %in% Type[[1]]))
# [1]  TRUE  TRUE  TRUE  TRUE FALSE

CodePudding user response：

Here's another possible option, where we can remove any colours from the list, then clean up the remaining characters (remove white space and remove commas). Then, I use nzchar to detect if there are any remaining words, if so, then remove that row.

Data[!nzchar(trimws(gsub(
  "[[:punct:]]", "", gsub(paste0(Type[[1]], collapse = "|"), "", Data$colour)
))), ]

Output

          colour
1           blue
2     blue,green
3 blue,black,red
4      black,red