Question

I have a data table of strings which I am trying to turn into lists and sort without sorting the vectors themselves:

> DT <- data.table(x=c("A11, A12, A41",
+                      "A11, A41, B11, A6, A21",
+                      "A41, A12",
+                      "A12, A41",
+                      "A12, A6, B41, A93"));DT
                        x
1:          A11, A12, A41
2: A11, A41, B11, A6, A21
3:               A41, A12
4:               A12, A41
5:      A12, A6, B41, A93

which I need to get to:

                        x
1:          A11, A12, A41
2: A11, A21, A41, A6, B11
3:               A12, A41
4:               A12, A41
5:      A12, A6, A93, B41

I’ve tried a using a bunch of ordering, as.lists, and in data.frames, as well as a data.table function I tried to write:

sortlists <- function(DT,col){
  for(i in 1:length(DT[,col])){
    DT[i,col]=order(DT[i,col])
  }
}

which throws an error that my column position is greater than ncol(x). Regardless, there must be a better way to get the result I want. I’m fairly new to R and am very new to data.tables (which I would like to impliment more since I have heard good things) so any help would be greatly appreciated!

Answer 1

This will split the character vectors then sort and return a list for each row

 DT[, list(y =  lapply(strsplit(x,', '), sort))]
                    y
1:        A11,A12,A41
2: A11,A21,A41,A6,B11
3:            A12,A41
4:            A12,A41
5:     A12,A6,A93,B41

And. if you really wanted the single character strings for each row

DT[, list(y =  vapply(strsplit(x,', '), function(x) paste(sort(x), collapse = ', '), FUN.VALUE =character(1)))]