colwise-select.md
In dplyr: A Grammar of Data Manipulation

colwise select() / rename() give meaningful errors

Code
  df <- tibble(x = 0L, y = 0.5, z = 1)
  (expect_error(df %>% rename_all()))
Output
  <error/rlang_error>
  Error in `rename_all()`:
  ! `.funs` must specify a renaming function.
Code
  (expect_error(df %>% rename_if(is_integerish)))
Output
  <error/rlang_error>
  Error in `rename_if()`:
  ! `.funs` must specify a renaming function.
Code
  (expect_error(df %>% rename_at(vars(x:y))))
Output
  <error/rlang_error>
  Error in `rename_at()`:
  ! `.funs` must specify a renaming function.
Code
  (expect_error(df %>% rename_all(list(tolower, toupper))))
Output
  <error/rlang_error>
  Error in `rename_all()`:
  ! `.funs` must contain one renaming function, not 2.
Code
  (expect_error(df %>% select_all(list(tolower, toupper))))
Output
  <error/rlang_error>
  Error in `select_all()`:
  ! `.funs` must contain one renaming function, not 2.
Code
  (expect_error(df %>% select_if(function(.x) 1)))
Output
  <error/rlang_error>
  Error in `select_if()`:
  ! `.p` is invalid.
  x `.p` should return a single logical.
  i `.p` returns a <double> for column `x`.
Code
  (expect_error(df %>% select_if(function(.x) c(TRUE, TRUE))))
Output
  <error/rlang_error>
  Error in `select_if()`:
  ! `.p` is invalid.
  x `.p` should return a single logical.
  i `.p` returns a size 2 <logical> for column `x`.
Code
  (expect_error(data.frame() %>% select_all(.funs = 42)))
Output
  <error/rlang_error>
  Error in `select_all()`:
  ! `.funs` must be a one sided formula, a function, or a function name.