Conditions raised by zuutf — zuutf

Every error zuutf raises inherits from zuutf_error, then error and condition, so a caller can catch all of them, or one kind, with tryCatch(). Test code should match on the class, never on the message.

Details

Class	Fields	Raised when
`zuutf_invalid_argument`		wrong type, unknown encoding name, a `"bytes"` string without `useBytes`
`zuutf_encoding_error`	`index`, `offset`, `encoding`	input that is ill-formed in its encoding under `invalid = "error"`, or a code point a Latin-1 target cannot represent
`zuutf_nul_error`	`index`, `offset`	`utf_decode()` meets U+0000, which an R string cannot hold
`zuutf_base64_error`	`offset`	Base64 input that `base64_decode()`'s mode rejects
`zuutf_size_error`		an output longer than a string (2^31 - 1 bytes) or a raw vector can hold
`zuutf_internal_error`		a simdutf result zuutf did not expect, or a kernel forced through `SIMDUTF_FORCE_IMPLEMENTATION` that this processor cannot run

index is the 1-based element of the input, offset the 1-based position of the first error in the input's code units (bytes for UTF-8, Latin-1 and ASCII; 16-bit units for UTF-16; 32-bit units for UTF-32), counted after any byte-order mark, and encoding the name of the encoding.

Examples

tryCatch(
  utf_valid("abc", encoding = "EBCDIC"),
  zuutf_invalid_argument = function(e) class(e)
)
#> [1] "zuutf_invalid_argument" "zuutf_error"            "error"                 
#> [4] "condition"