Skip to contents

Try to coerce or validate x as all of several specs

Source: R/stabilize_all_of.R
stabilize_all_of.Rd

stabilize_all_of() validates and coerces x by applying every function in ... to x, independently. x must satisfy every spec, and all specs must agree on the coerced result; if any spec fails, or specs disagree on the coerced value, an informative error is thrown. stabilise_all_of() is a synonym.

Usage

stabilize_all_of(
  x,
  ...,
  x_arg = caller_arg(x),
  call = caller_env(),
  x_class = object_type(x)
)

stabilise_all_of(
  x,
  ...,
  x_arg = caller_arg(x),
  call = caller_env(),
  x_class = object_type(x)
)

Arguments

x

The object to stabilize.

...

Unnamed stabilizer functions, such as stabilize_* functions (stabilize_chr(), etc.), to_* functions (to_chr(), etc.), functions produced by specify_*() calls (specify_chr(), etc.), or assert_*() functions (such as assert_not()) that return their input unchanged. Each is applied to the original x; x must pass every one of them, and they must all return the same value.

x_arg

(character(1)) The name of the object being stabilized to use in error messages. The automatic value will work in most cases, or pass it through from higher-level functions to make error messages clearer in unexported functions.

call

(environment) The execution environment to mention as the source of error messages.

x_class

(character(1)) The class name of the object being stabilized to use in error messages. Use this if you remove a special class from the object before checking its coercion, but want the error message to match the original class.

Value

x coerced or validated by every function in ..., or an error condition with classes <stbl-error>, <stbl-condition>, <rlang_error>, <error>, <condition>, and a specific class by failure mode:

  • <stbl-error-empty_specs> when no functions are supplied in ....

  • <stbl-error-named_spec> when any element of ... is named.

  • <stbl-error-cant_stabilize_all_of> when any provided function fails.

  • <stbl-error-inconsistent_all_of> when every function succeeds, but they don't all produce the same coerced value.

See also

Other stabilization functions: assert_contains(), assert_not(), assert_present(), stabilize_any_of(), stabilize_arg(), stabilize_chr(), stabilize_chr_scalar(), stabilize_date(), stabilize_date_scalar(), stabilize_dbl(), stabilize_dbl_scalar(), stabilize_df(), stabilize_dttm(), stabilize_dttm_scalar(), stabilize_dur(), stabilize_dur_scalar(), stabilize_fct(), stabilize_fct_scalar(), stabilize_int(), stabilize_int_scalar(), stabilize_lgl(), stabilize_lgl_scalar(), stabilize_lst(), stabilize_one_of(), stabilize_time(), stabilize_time_scalar(), to_chr(), to_chr_scalar(), to_date(), to_date_scalar(), to_dbl(), to_dbl_scalar(), to_dttm(), to_dttm_scalar(), to_dur(), to_dur_scalar(), to_fct(), to_fct_scalar(), to_int(), to_int_scalar(), to_lgl(), to_lgl_scalar(), to_time(), to_time_scalar()

Examples

# Returns x unchanged when all functions succeed
stabilize_all_of(1L, stabilize_int, stabilize_int_scalar)
#> [1] 1

# Each spec is applied to the original x, independently
stabilize_all_of("ab", specify_chr(regex = "^a"), specify_chr(regex = "b$"))
#> [1] "ab"

# Errors with a combined message when a spec fails
try(stabilize_all_of("a", stabilize_int, stabilize_chr))
#> Error in eval(expr, envir) : 
#>   `"a"` must match all of the provided stabilizers.
#> ✖ `"a"` <character> must be coercible to <integer> (Locations: 1)

# Errors because each spec sees the original "1" (a string); the second
# spec would succeed on 1L, but doesn't get the chance to see it
try(stabilize_all_of("1", stabilize_int, specify_dbl(coerce_character = FALSE)))
#> Error in eval(expr, envir) : 
#>   `"1"` must match all of the provided stabilizers.
#> ✖ Can't coerce `"1"` <character> to <double>.

# Errors when specs succeed but disagree on the coerced value
# (stabilize_int() keeps 1L an integer; specify_dbl() makes it a double)
try(stabilize_all_of(1L, stabilize_int, specify_dbl()))
#> Error in eval(expr, envir) : 
#>   `1L` must be coerced the same way by every provided stabilizer.
#> ℹ The provided stabilizers succeeded, but disagreed on the coerced value.