`99ab201`

prioritize fuzzy match quality over frecency

Authored by mfwolffe <wolffemf@dukes.jmu.edu> 3 months ago

Status	File	+	-
M	`Cargo.lock`	1	1
M	`src/matcher/fuzzy.rs`	7	5

Cargo.lockmodified

  [[package]]
  name = "gump"
 -version = "0.2.1"
 +version = "0.2.2"
  dependencies = [
   "bincode",
   "chrono",

src/matcher/fuzzy.rsmodified

              .filter_map(|(path, frecency)| {
                  let fuzzy_score = self.score(path, terms)?;
 -                // Combined score: frecency is the primary factor, fuzzy score is secondary
 -                // Frecency typically ranges 0.25-40+, fuzzy score 0-500+
 -                // We normalize fuzzy score to 0-1 range and use it as a multiplier
 -                let fuzzy_factor = 1.0 + (fuzzy_score as f64 / 500.0).min(1.0);
 -                let combined_score = frecency * fuzzy_factor;
 +                // Combined score: fuzzy match quality is PRIMARY, frecency is secondary
 +                // This ensures a great match beats a poor match with high frecency
 +                //
 +                // Fuzzy score (0-500+) is the base
 +                // Frecency (0.25-40+) adds a small boost (up to ~25%) for tiebreaking
 +                let frecency_bonus = 1.0 + (frecency / 160.0).min(0.25);
 +                let combined_score = (fuzzy_score as f64) * frecency_bonus;
                  Some(Match {
                      path,