technofab/pay-respects
1
0
Fork 0
mirror of https://github.com/TECHNOFAB11/pay-respects.git synced 2025-12-12 14:30:10 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: use damerau variation for string comparison

Browse source
This commit is contained in:
iff 2025-04-09 15:50:37 +02:00
parent f31e508696
commit b2a01a97db
1 changed files with 10 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

10
utils/src/evals.rs
View file

@ -181,6 +181,7 @@ pub fn find_similars(
None None
} }
/// Damerau-Levenshtein distance algorithm
#[allow(clippy::needless_range_loop)] #[allow(clippy::needless_range_loop)]
pub fn compare_string(a: &str, b: &str) -> usize { pub fn compare_string(a: &str, b: &str) -> usize {
let mut matrix = vec![vec![0; b.chars().count() + 1]; a.chars().count() + 1]; let mut matrix = vec![vec![0; b.chars().count() + 1]; a.chars().count() + 1];
@ -199,6 +200,15 @@ pub fn compare_string(a: &str, b: &str) -> usize {
std::cmp::min(matrix[i][j + 1] + 1, matrix[i + 1][j] + 1), std::cmp::min(matrix[i][j + 1] + 1, matrix[i + 1][j] + 1),
matrix[i][j] + cost, matrix[i][j] + cost,
); );
// addition for optimal string alignment distance
if i > 0
&& j > 0 && ca == b.chars().nth(j - 1).unwrap()
&& a.chars().nth(i - 1).unwrap() == cb
{
matrix[i + 1][j + 1] =
std::cmp::min(matrix[i + 1][j + 1], matrix[i - 1][j - 1] + 1);
}
} }
} }
matrix[a.chars().count()][b.chars().count()] matrix[a.chars().count()][b.chars().count()]