Question

I would like to determine whether a list contains any duplicate elements, while considering permutations as equivalent. All vectors are of equal length.

What is the most efficient way (shortest running time) to accomplish this?

## SAMPLE DATA
a  <- c(1, 2, 3)
b  <- c(4, 5, 6)
a.same <- c(3, 1, 2)

## BOTH OF THSE LISTS SHOULD BE FLAGGED AS HAVING DUPLICATES
myList1 <- list(a, b, a)
myList2 <- list(a, b, a.same)


# CHECK FOR DUPLICATES
anyDuplicated(myList1) > 0  # TRUE
anyDuplicated(myList2) > 0  # FALSE, but would like true. 

For now I am resorting to sorting each member of the list before checking for duplicates

anyDuplicated( lapply(myList2, sort) ) > 0

I am wondering if there is a more efficient alternative. Also, in the ?duplicated documentation, it indicates “Using this for lists is potentially slow”. Are there other functions better suited for lists?

Answer 1

What about this…?

a  <- c(1, 2, 3)
b  <- c(4, 5, 6)
a.same <- c(3, 1, 2)
myList1 <- list(a, b, a)
myList2 <- list(a, b, a.same)

# For exact duplicated values: List1
DF1 <- do.call(rbind, myList1)  # From list to data.frame
ind1 <- apply(DF1, 2, duplicated) # logical matrix for duplicated values
DF1[ind1] # finding duplicated values  
[1] 1 2 3

# For permutations: List2
DF2 <- do.call(rbind, myList2)
ind2 <- apply(apply(DF2, 1, sort), 1, duplicated)
DF2[ind2] # duplicated values
[1] 3 1 2