CODEWITHSUNDEEP
r
wendigo
wendigo

Reputation: 33

sample() gives duplicate results - how to avoid?

I have multiple functions following this schematic:

function1 <- function(){
    string_samples <- c("string1", "string2", "string3", "string4", "string5")
sample(string_samples, 1)
}

function1 gets called in another function as follows:

function_process <- function(){
a <- sample(1:3,1)
a2 <- sample(1:3,1)

g <- gsub(" ", "", paste(function1(), "-", sample(1:a,1)))
g2 <- gsub(" ", "", paste(function1(), "-", sample(1:a2,1)))

rnd_g <- sample(1:2, 1)
if(rnd_g==1) x <- paste(g)
else if(rnd_g==2) x <- paste(g, g2)

  return(x)
}

I want to avoid that one string value is getting picked multiple times at return(x). How could I achieve this in a simple way?

Right now results could be something like: string1-1 string1-2 which I want to avoid.

Upvotes: 2

Views: 44

Answers (1)

Allan Cameron
Allan Cameron

Reputation: 173898

This seems like a strange way to do things, but the following simplified function will achieve it:

function_process <- function(){
  
  a  <- sample(1:3, 1)
  a2 <- sample(1:3, 1)
  
  g  <- function1()
  g2 <- function1()
  
  while(g == g2) {
    g2 <- function1()
  }

  g  <- paste0(g, "-", sample(1:a,1))
  g2 <- paste0(g2, "-", sample(1:a2, 1))
  
  if(sample(0:1, 1)) return(g)
  
  paste(g, g2)
}

Note that you can use paste0 instead of using gsub to remove the spaces that paste adds by default.

You can see the same string is never used twice here:

set.seed(1)

for(i in 1:10) print(function_process())
#> [1] "string1-1"
#> [1] "string3-1"
#> [1] "string1-1"
#> [1] "string2-1"
#> [1] "string2-2 string4-2"
#> [1] "string4-1"
#> [1] "string2-2"
#> [1] "string3-1 string4-3"
#> [1] "string1-1"
#> [1] "string4-1"

Created on 2022-01-31 by the reprex package (v2.0.1)

Upvotes: 1

Related Questions