How to bypass a missing argument in r

Question

I wrote a program that takes more than one vector as arguments. If I input a single vector as an input the function provides the following error message:

Error in combn(letters[1:length(mylist)], 2) : n < m
Called from: combn(letters[1:length(mylist)], 2)

I know why I am getting this error. I have used a combination function combn that chooses 2 vectors. How do I fix this? I want to get an output even if my input is a single vector. In this case, I want my code to skip those lines that need 2 vectors and run the rest to get an output. I am providing my entire code here for convenience.

### Function to check whether two sets a,b are incomparable

incomparable <- function(a,b){
  
  cond1 <- sum(a>b) > 0 # Returns true is if there is at least one case such that a_i is 1 and b_i is 0
  cond2 <- sum(b>a) > 0 # True if at least one case such that one case with b_i is 1 and a_i is 0
  ifelse (length(a)== length(b), cond1*cond2, "Strings are not of the same length")
}



incomparable(a,b)

### Function to get subsets of a binary vector

binary_subset<-function(a){
  a_seq = lapply(a, seq, 0)   # keep 0s as 0, make 1s c(1, 0)
  subset=do.call(expand.grid, a_seq)
  colnames(subset)=(1:length(a))
  return(subset)
} 

#### Function to generate all lower-order interaction terms correspond to interaction terms a,b,c...upto any arbitrary number L (including terms a, b, c...L)

all_lower_order_interactions<-function(...){
  mylist <- list(...)

  combination<-combn(letters[1:length(mylist)], 2) #choose 2 out of L vectors
  
  check_incomparable<-0
  for (j in 1:ncol(combination)){
    check_incomparable[j]<- (incomparable(get(combination[1,j]), get(combination[2,j])))
  }
  check_incomparable
  
  if(all(check_incomparable>0)==FALSE) {stop( "at least one of the interaction terms is a special case (or a subset) of another term.")}
  
  interactions_abc <- do.call("rbind", lapply(mylist, binary_subset))
  interactions_no_duplicate <- unique(interactions_abc[1:length(mylist[[1]])])
  rownames(interactions_no_duplicate) <- 1:nrow(interactions_no_duplicate)
  
  interactions_no_duplicate
}



a<-c(0,1,0)
b<-c(0,0,1)
all_lower_order_interactions(a)

What exactly are the lines that need 2 vectors? Only combination<-combn(letters[1:length(mylist)], 2)? — starja

Waldi Waldi · Accepted Answer · 2020-09-20T19:38:30

You could use an if statement to skip the lines depending on combn in the case there are not at least 2 vectors:

all_lower_order_interactions <- function(...) {
  mylist <- list(...)
  if (length(mylist) >= 2) {
    {
      combination <-
        combn(letters[1:length(mylist)], 2) #choose 2 out of L vectors
    }
    
    check_incomparable <- 0
    for (j in 1:ncol(combination)) {
      check_incomparable[j] <-
        (incomparable(get(combination[1, j]), get(combination[2, j])))
    }
    check_incomparable
    
    if (all(check_incomparable > 0) == FALSE) {
      stop("at least one of the interaction terms is a special case (or a subset) of another term.")
    }
  }
  interactions_abc <-
    do.call("rbind", lapply(mylist, binary_subset))
  interactions_no_duplicate <-
    unique(interactions_abc[1:length(mylist[[1]])])
  rownames(interactions_no_duplicate) <-
    1:nrow(interactions_no_duplicate)
  
  interactions_no_duplicate
}


a <- c(0, 1, 0)
b <- c(0, 0, 1)
all_lower_order_interactions(a)

  1 2 3
1 0 1 0
2 0 0 0

How to bypass a missing argument in r

2 Answers

Edit