I have two datasets as the ones described below:

dfA <- tibble(
  name = c("John", "Michael", "Brian", "Thomas", "Peter"),
  expected = c(128.34, 453.6674, 789.2345, 678.34, 475.467),
  max = c(128.3464, 453.6901, 789.2740, 678.3739, 475.4908),
  min = c(128.3336, 453.6447, 789.1950, 678.3061, 475.4432)
)

dfB <- tibble(
  real = c(128.76, 453.658, 789.235, 677.2035, 475.4695),
  time = c(3.24, 5.65, 4.89, 2.28, 6.12)
)

dfA$max and dfA$min have been previously calculated using the following functions:

shift <- function (theoretical, tolerance) {
  return(tolerance*theoretical/10^6)
}

calculate_max <- function (vector, tolerance) {
  res <- c()
  for (i in vector) {
    max <- i + shift(i, tolerance)
    res <- append(res, max)
  }
  return(res)
}

calculate_min <- function (vector, tolerance) {
  res <- c()
  for (i in vector) {
    min <- i - shift(i, tolerance)
    res <- append(res, min)
  }
  return(res)
}

I’d like to intersect both datasets so that dfB$real column lies between dfA$min and dfA$max columns.

I have already found a solution using the fuzzyjoin library, but I’m intrigued about how to do this manually, as iterating over rows with tibbles does not seem to be a good practice from what I’ve read and I cannot think of another way of achieving this without looping over rows.

The working solution with fuzzyjoin would be as follows (in case it’s helpful to anyone):

fuzzy_right_join(dfB, dfA,
                 by = c("real"="max", "real"="min"),
                 match_fun = list(`<=`, `>=`))

Thanks for your help!