use unicode_normalization::UnicodeNormalization;
use unicode_normalization::char::is_combining_mark;
#[cfg(test)]
pub(crate) fn normalize_nfc(text: &str) -> String {
text.nfc().collect()
}
pub(crate) fn normalize_nfkc(text: &str) -> String {
text.nfkc().collect()
}
pub(crate) fn case_fold(text: &str) -> String {
text.chars().flat_map(|c| c.to_lowercase()).collect()
}
pub(crate) fn accent_fold(text: &str) -> String {
text.nfd()
.filter(|c| !is_combining_mark(*c))
.nfc()
.collect()
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn nfc_recomposes_decomposed_acute() {
let decomposed = "cafe\u{0301}";
assert_eq!(normalize_nfc(decomposed), "café");
}
#[test]
fn nfc_passes_already_composed() {
assert_eq!(normalize_nfc("café"), "café");
}
#[test]
fn nfkc_compatibility_decomposes_superscript() {
assert_eq!(normalize_nfkc("x²"), "x2");
}
#[test]
fn nfkc_compatibility_decomposes_ligature() {
assert_eq!(normalize_nfkc("final"), "final");
}
#[test]
fn case_fold_lowercases_ascii_and_latin() {
assert_eq!(case_fold("Hello WORLD"), "hello world");
assert_eq!(case_fold("Café"), "café");
}
#[test]
fn case_fold_handles_greek() {
assert_eq!(case_fold("Ω"), "ω");
}
#[test]
fn accent_fold_strips_latin_diacritics() {
assert_eq!(accent_fold("café"), "cafe");
assert_eq!(accent_fold("naïve"), "naive");
assert_eq!(accent_fold("résumé"), "resume");
}
#[test]
fn accent_fold_preserves_unaccented_text() {
assert_eq!(accent_fold("hello world"), "hello world");
}
#[test]
fn accent_fold_handles_multiple_combining_marks() {
let multi = "a\u{0301}\u{0308}";
assert_eq!(accent_fold(multi), "a");
}
#[test]
fn accent_fold_handles_decomposed_input() {
let decomposed = "cafe\u{0301}";
assert_eq!(accent_fold(decomposed), "cafe");
}
}