Merge 9ece62d into 81e5320

rrriiiccckkk · web-flow · commit f957ad7797b9 · 2026-06-26T14:59:57.000+08:00
diff --git a/extensions/memory-core/src/memory/hybrid.ts b/extensions/memory-core/src/memory/hybrid.ts
@@ -49,6 +49,13 @@ export function bm25RankToScore(rank: number): number {
   return 1 / (1 + rank);
 }
 
+/**
+ * Check if two line ranges overlap.
+ */
+function linesOverlap(aStart: number, aEnd: number, bStart: number, bEnd: number): boolean {
+  return aStart <= bEnd && bStart <= aEnd;
+}
+
 export async function mergeHybridResults(params: {
   vector: HybridVectorResult[];
   keyword: HybridKeywordResult[];
@@ -87,6 +94,9 @@ export async function mergeHybridResults(params: {
     }
   >();
 
+  // Secondary index: group vector results by path for fallback matching
+  const vectorByPath = new Map<string, HybridVectorResult[]>();
+
   for (const r of params.vector) {
     byId.set(r.id, {
       id: r.id,
@@ -98,26 +108,64 @@ export async function mergeHybridResults(params: {
       vectorScore: r.vectorScore,
       textScore: 0,
     });
+    const existing = vectorByPath.get(r.path);
+    if (existing) {
+      existing.push(r);
+    } else {
+      vectorByPath.set(r.path, [r]);
+    }
   }
 
   for (const r of params.keyword) {
     const existing = byId.get(r.id);
     if (existing) {
+      // Exact chunk ID match — update textScore directly
       existing.textScore = r.textScore;
       if (r.snippet && r.snippet.length > 0) {
         existing.snippet = r.snippet;
       }
     } else {
-      byId.set(r.id, {
-        id: r.id,
-        path: r.path,
-        startLine: r.startLine,
-        endLine: r.endLine,
-        source: r.source,
-        snippet: r.snippet,
-        vectorScore: 0,
-        textScore: r.textScore,
-      });
+      // No chunk ID match — try path + line range overlap as fallback
+      // Pick the overlapping vector chunk with the highest vectorScore
+      const candidates = vectorByPath.get(r.path);
+      let merged = false;
+      if (candidates) {
+        let best: HybridVectorResult | undefined;
+        for (const v of candidates) {
+          if (linesOverlap(r.startLine, r.endLine, v.startLine, v.endLine)) {
+            if (!best || v.vectorScore > best.vectorScore) {
+              best = v;
+            }
+          }
+        }
+        if (best) {
+          const entry = byId.get(best.id)!;
+          entry.textScore = Math.max(entry.textScore, r.textScore);
+          if (r.startLine < entry.startLine) {
+            entry.startLine = r.startLine;
+          }
+          if (r.endLine > entry.endLine) {
+            entry.endLine = r.endLine;
+          }
+          if (r.snippet && r.snippet.length > 0) {
+            entry.snippet = r.snippet;
+          }
+          merged = true;
+        }
+      }
+      if (!merged) {
+        // No overlap found — add as keyword-only entry
+        byId.set(r.id, {
+          id: r.id,
+          path: r.path,
+          startLine: r.startLine,
+          endLine: r.endLine,
+          source: r.source,
+          snippet: r.snippet,
+          vectorScore: 0,
+          textScore: r.textScore,
+        });
+      }
     }
   }
 
diff --git a/extensions/memory-core/src/memory/manager-search.ts b/extensions/memory-core/src/memory/manager-search.ts
@@ -123,7 +123,7 @@ function planKeywordSearch(params: {
   const matchTerms: string[] = [];
   const substringTerms: string[] = [];
   for (const token of tokens) {
-    if (SHORT_CJK_TRIGRAM_RE.test(token) && Array.from(token).length < 3) {
+    if (SHORT_CJK_TRIGRAM_RE.test(token)) {
       substringTerms.push(token);
       continue;
     }
@@ -380,6 +380,27 @@ export async function searchKeyword(params: {
           params.limit,
         ) as typeof rows;
       usedMatch = true;
+      // If MATCH returned 0 rows, fall back to LIKE for better recall
+      if (rows.length === 0) {
+        const queryTokens = normalizeStringEntries(params.query.match(FTS_QUERY_TOKEN_RE) ?? []);
+        const allTerms = uniqueStrings([...queryTokens, ...plan.substringTerms]);
+        const fbLike = allTerms.map(() => " OR text LIKE ? ESCAPE '\\'").join("");
+        const fbParams = allTerms.map((term) => `%${escapeLikePattern(term)}%`);
+        rows = params.db
+          .prepare(
+            `SELECT id, path, source, start_line, end_line, text,
+` +
+              `       0 AS rank
+` +
+              `  FROM ${params.ftsTable}
+` +
+              ` WHERE 1=0${fbLike}${liveChunkClause}${params.sourceFilter.sql}
+` +
+              ` LIMIT ?`,
+          )
+          .all(...fbParams, ...params.sourceFilter.params, params.limit) as typeof rows;
+        usedMatch = false;
+      }
     } catch (matchErr) {
       // FTS5 MATCH can fail on certain token patterns depending on the
       // Node.js sqlite runtime and tokenizer (e.g. unicode61 vs trigram).
