From 83d3f4bd2140e577dee856488325b30124162f72 Mon Sep 17 00:00:00 2001
From: r266-tech <r266-tech@users.noreply.github.com>
Date: Fri, 30 Jan 2026 14:56:10 +0800
Subject: [PATCH] fix: support Unicode characters in hybrid search FTS query

The previous regex /[A-Za-z0-9_]+/g only matched ASCII characters,
causing CJK (Chinese/Japanese/Korean) and other non-ASCII Unicode
characters to be completely ignored during hybrid search.

Changed to /[\p{L}\p{N}_]+/gu which uses Unicode property escapes:
- \p{L} matches all Unicode letters (including CJK)
- \p{N} matches all Unicode numbers
- u flag enables Unicode mode

This ensures hybrid search works correctly with Chinese, Japanese,
Korean, and other non-Latin scripts.
---
 src/memory/hybrid.ts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/memory/hybrid.ts b/src/memory/hybrid.ts
index 753748bf9..0f86f4c6d 100644
--- a/src/memory/hybrid.ts
+++ b/src/memory/hybrid.ts
@@ -23,7 +23,7 @@ export type HybridKeywordResult = {
 export function buildFtsQuery(raw: string): string | null {
   const tokens =
     raw
-      .match(/[A-Za-z0-9_]+/g)
+      .match(/[\p{L}\p{N}_]+/gu)
       ?.map((t) => t.trim())
       .filter(Boolean) ?? [];
   if (tokens.length === 0) return null;