Creating resuable stub in unit testing?

Question

I am looking for a way to reuse some stubs and mocks. I thought of encapsulating them inside a function. The advantage would be that I could modify that one function instead of modifying every single instance a function is being stubbed out separately. I think the issue has to do with how stub deals with the test environment. I am using mockery::stub.

Some example would be:

 stub_some_function <- function(parent_function){
    mockery::stub(
        where = parent_function,
        what = "content",
        how = function(x, ...) TRUE
    )
  }

I could then have a test with a loop that calls the stubs I need (or do it inside a function and call that function):

add_2 <- function(x) x + 2
get_result <- function(x) add_2(x)

test_that("Returns TRUE", {
  stub_some_function(get_result)
  
  expect_true(get_result(4))

})

or with more stubs

add_2 <- function(x) x + 2
remove_4 <- function(x) x - 4

get_result <- function(x) {
  res <- add_2(x)
  remove_4(res)
}

stub_add_2 <- function(parent_function){
    mockery::stub(
        where = parent_function,
        what = "add_2",
        how = function(x, ...) TRUE
    )
}

stub_remove_4 <- function(parent_function){
    mockery::stub(
        where = parent_function,
        what = "remove_4",
        how = function(x, ...) FALSE
    )
}

stub_multiple(parent, ...){
  dots <- list(...)
  for (func in dots){
    func(parent)
  }
}

test_that("Returns FALSE", {
  stub_multiple(get_result, stub_add_2, stub_remove_4)

  expect_false(get_result(4)) # returns 2
})

Or even if I would call the stub_function(the_parent) inline as in:

test_that("Returns FALSE", {
  stub_add_2(get_result)
  stub_remove_4(get_result)

  expect_false(get_result(4)) # returns 2
})

The test fails because the functions are not stubbed out. I think one solution would be to have the whole test (test_that) inside a function itself so that mockery uses the test environment.