Evaluate an R expression with a temporarily set random set

Evaluate an R expression with a temporarily set random set.

withSeed(expr, seed, ..., substitute=TRUE, envir=parent.frame())

Arguments

expr

The R expression to be evaluated.

seed, ...

Arguments passed to set.seed().

substitute

If TRUE, argument expr is substitute():ed, otherwise not.

envir

The environment in which the expression should be evaluated.

Value

Returns the results of the expression evaluated.

Details

Upon exit (also on errors), this function will restore .Random.seed in the global environment to the value it had upon entry. If it did not exist, it will be removed.

Author

Henrik Bengtsson

See also

Internally, set.seed() is used to set the random seed.

Examples

# Generate a random number
y0 <- runif(1)
print(y0)
#> [1] 0.5637715

# Generate a random number using the same seed over and over
yp <- NULL
for (ii in 1:10) {
  y <- withSeed({
    runif(1)
  }, seed=0x42)
  print(y)
  # Assert identical
  if (!is.null(yp)) stopifnot(identical(y, yp))
  yp <- y
}
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366
#> [1] 0.9899366

# Generate a random number
y <- runif(1)
print(y)
#> [1] 0.1299701