tests/testthat/_snaps/if-else.md

In dplyr: A Grammar of Data Manipulation

takes the common type of true and false (#6243)

Code
  if_else(TRUE, 1, "x")
Condition
  Error in `if_else()`:
  ! Can't combine `true` <double> and `false` <character>.

includes missing in the common type computation if used

Code
  if_else(TRUE, 1, 2, missing = "x")
Condition
  Error in `if_else()`:
  ! Can't combine `true` <double> and `missing` <character>.

condition must be logical (and isn't cast to logical!)

Code
  if_else(1:10, 1, 2)
Condition
  Error in `if_else()`:
  ! `condition` must be a logical vector, not an integer vector.

condition can't be an array (#7723)

Code
  if_else(array(TRUE), 1, 2)
Condition
  Error in `if_else()`:
  ! `condition` must be a logical vector, not a logical 1D array.

true, false, and missing must recycle to the size of condition

Code
  if_else(x < 2, bad, x)
Condition
  Error in `if_else()`:
  ! Can't recycle `true` (size 2) to size 3.

Code
  if_else(x < 2, x, bad)
Condition
  Error in `if_else()`:
  ! Can't recycle `false` (size 2) to size 3.

Code
  if_else(x < 2, x, x, missing = bad)
Condition
  Error in `if_else()`:
  ! Can't recycle `missing` (size 2) to size 3.

must have empty dots

Code
  if_else(TRUE, 1, 2, missing = 3, 4)
Condition
  Error in `if_else()`:
  ! `...` must be empty.
  x Problematic argument:
  * ..1 = 4
  i Did you forget to name an argument?

ptype overrides the common type

Code
  if_else(TRUE, 1L, 2.5, ptype = integer())
Condition
  Error in `if_else()`:
  ! Can't convert from `false` <double> to <integer> due to loss of precision.
  * Locations: 1

size is deprecated

Code
  x <- if_else(c(TRUE, FALSE), 1, 2, size = 2)
Condition
  Warning:
  The `size` argument of `if_else()` is deprecated as of dplyr 1.2.0.

dplyr documentation built on Feb. 3, 2026, 9:08 a.m.