45 lines
1.9 KiB
Rust
45 lines
1.9 KiB
Rust
use icu::{
|
|
collator::{AlternateHandling, Collator, CollatorOptions, MaxVariable, Strength},
|
|
segmenter::GraphemeClusterSegmenter,
|
|
};
|
|
use itertools::Itertools;
|
|
use std::collections::HashSet;
|
|
|
|
// We define an anagram as any permutation of the graphemes of a string, ignoring
|
|
// case, accents/diacritical marks, punctuation, spaces, etc. There are some
|
|
// language-specific instances where this fails, but it covers a lot of cases.
|
|
pub fn anagrams_for<'a>(word: &str, possible_anagrams: &'a [&str]) -> HashSet<&'a str> {
|
|
let mut options = CollatorOptions::new();
|
|
// Ignore case sensitivity, accents, punctuation, spaces, etc.
|
|
options.strength = Some(Strength::Primary);
|
|
options.alternate_handling = Some(AlternateHandling::Shifted);
|
|
options.max_variable = Some(MaxVariable::Punctuation);
|
|
// Strictly speaking an anagram only makes sense within a given locale,
|
|
// but we don't know it, so leave it as undefined.
|
|
let collator = Collator::try_new(Default::default(), options).unwrap();
|
|
let word_permutation = first_grapheme_permutation(&word, &collator);
|
|
possible_anagrams
|
|
.iter()
|
|
.copied()
|
|
.filter(|possibility| {
|
|
let possibility_permutation = first_grapheme_permutation(&possibility, &collator);
|
|
// If two strings are anagrams of each other, they will have the same first
|
|
// permutation.
|
|
let is_anagram = collator
|
|
.compare(&word_permutation, &possibility_permutation)
|
|
.is_eq();
|
|
let is_same = collator.compare(&possibility, &word).is_eq();
|
|
is_anagram && !is_same
|
|
})
|
|
.collect()
|
|
}
|
|
|
|
fn first_grapheme_permutation(word: &str, collator: &Collator) -> String {
|
|
let segmenter = GraphemeClusterSegmenter::new();
|
|
segmenter
|
|
.segment_str(&word)
|
|
.tuple_windows()
|
|
.map(|(i, j)| &word[i..j])
|
|
.sorted_by(|a, b| collator.compare(a, b))
|
|
.collect()
|
|
}
|