@dicelette/core 1.26.0 → 1.27.0

package/dist/index.d.mts

@@ -318,7 +318,7 @@ declare function calculateSimilarity(str1: string, str2: string): number;
  * Calculates the Levenshtein distance between two strings.
  */
 declare function levenshteinDistance(str1: string, str2: string): number;
-declare function findBestStatMatch<T>(searchTerm: string, normalizedStats: Map<string, T>, similarityThreshold?: number, partialSearch?: boolean): T | undefined;
+declare function findBestStatMatch<T>(searchTerm: string, normalizedStats: Map<string, T>, similarityThreshold?: number): T | undefined;
 /**
  * Find the snippet name with the highest similarity to `macroName`.
  * Single-pass O(n) algorithm: keeps the best (name, similarity) seen so far.

package/dist/index.d.ts

@@ -318,7 +318,7 @@ declare function calculateSimilarity(str1: string, str2: string): number;
  * Calculates the Levenshtein distance between two strings.
  */
 declare function levenshteinDistance(str1: string, str2: string): number;
-declare function findBestStatMatch<T>(searchTerm: string, normalizedStats: Map<string, T>, similarityThreshold?: number, partialSearch?: boolean): T | undefined;
+declare function findBestStatMatch<T>(searchTerm: string, normalizedStats: Map<string, T>, similarityThreshold?: number): T | undefined;
 /**
  * Find the snippet name with the highest similarity to `macroName`.
  * Single-pass O(n) algorithm: keeps the best (name, similarity) seen so far.

package/dist/index.js

@@ -275,23 +275,18 @@ function levenshteinDistance(str1, str2) {
   }
   return matrix[str2.length][str1.length];
 }
-function findBestStatMatch(searchTerm, normalizedStats, similarityThreshold = MIN_THRESHOLD_MATCH, partialSearch = true) {
+function findBestStatMatch(searchTerm, normalizedStats, similarityThreshold = MIN_THRESHOLD_MATCH) {
   const exact = normalizedStats.get(searchTerm);
   if (exact) return exact;
-  if (partialSearch) {
-    const candidates = [];
-    for (const [normalizedKey, original] of normalizedStats) {
-      if (normalizedKey.startsWith(searchTerm))
-        candidates.push([original, normalizedKey.length]);
-      else if (normalizedKey.endsWith(searchTerm))
-        candidates.push([original, normalizedKey.length]);
-      else if (normalizedKey.includes(searchTerm))
-        candidates.push([original, normalizedKey.length]);
-    }
-    if (candidates.length > 0) {
-      candidates.sort((a, b) => a[1] - b[1]);
-      return candidates[0][0];
-    }
+  const candidates = [];
+  for (const [normalizedKey, original] of normalizedStats) {
+    if (normalizedKey.startsWith(searchTerm))
+      candidates.push([original, calculateSimilarity(searchTerm, normalizedKey)]);
+  }
+  if (candidates.length === 1) return candidates[0][0];
+  if (candidates.length > 0) {
+    candidates.sort((a, b) => b[1] - a[1]);
+    if (candidates[0][1] >= similarityThreshold) return candidates[0][0];
   }
   let bestMatch;
   let bestSimilarity = 0;
@@ -313,8 +308,7 @@ function findBestRecord(record, searchTerm, similarityThreshold = MIN_THRESHOLD_
   return findBestStatMatch(
     searchTerm.standardize(),
     normalizeRecord,
-    similarityThreshold,
-    false
+    similarityThreshold
   ) || null;
 }
 
@@ -333,7 +327,7 @@ function handleDiceAfterD(tokenStd, normalizedStats) {
   if (!diceMatch) return null;
   const diceCount = diceMatch[1] || "";
   const afterD = diceMatch[2];
-  const bestMatch = findBestStatMatch(afterD, normalizedStats, 1, false);
+  const bestMatch = findBestStatMatch(afterD, normalizedStats, 1);
   if (bestMatch) {
     const [, value] = bestMatch;
     return `${diceCount}d${value.toString()}`;
@@ -341,7 +335,7 @@ function handleDiceAfterD(tokenStd, normalizedStats) {
   return null;
 }
 function handleSimpleToken(tokenStd, token, normalizedStats, minThreshold) {
-  const bestMatch = findBestStatMatch(tokenStd, normalizedStats, minThreshold, false);
+  const bestMatch = findBestStatMatch(tokenStd, normalizedStats, minThreshold);
   if (bestMatch) {
     const [, value] = bestMatch;
     return value.toString();