pub fn uuid_normalise(text: &str) -> Result<String, String>Expand description
A UUID in the canonical 8-4-4-4-12 form, lowercased, or why it is not one.
Normalising rather than only validating is the whole reason this is a function and not a
str_len check in Beck: two spellings of the same UUID must not be two map keys. The braced and
unhyphenated forms other systems emit are read and written back canonically, so a program
comparing identifiers is comparing identity rather than punctuation.