From 63c2da2d3864bc821773e981de1285ae6bb8f39a Mon Sep 17 00:00:00 2001
From: Merge Sim <sim@local>
Date: Thu, 10 Sep 2026 01:11:39 -0700
Subject: [PATCH] refactor(ai-vault): move the surrogate-safe slice to shared,
 one implementation

`sliceAtCodeUnitLimit` lived in src/main/ai-vault, and src/shared never imports
src/main, so a shared consumer could not reach it. Rather than add a second
copy, it moves to src/shared and ai-vault imports and re-exports it, leaving
every existing importer of that module untouched.

Separated from the feature that needs it: this is the only change here to a
subsystem the rest of the branch does not touch.
---
 .../session-scanner-text-normalization.ts     | 13 ++++---------
 src/shared/surrogate-safe-text-slice.ts       | 19 +++++++++++++++++++
 2 files changed, 23 insertions(+), 9 deletions(-)
 create mode 100644 src/shared/surrogate-safe-text-slice.ts

diff --git a/src/main/ai-vault/session-scanner-text-normalization.ts b/src/main/ai-vault/session-scanner-text-normalization.ts
index 99f73a93d02..50fc54ab7aa 100644
--- a/src/main/ai-vault/session-scanner-text-normalization.ts
+++ b/src/main/ai-vault/session-scanner-text-normalization.ts
@@ -1,3 +1,7 @@
+import { sliceAtCodeUnitLimit } from '../../shared/surrogate-safe-text-slice'
+
+export { sliceAtCodeUnitLimit }
+
 const SESSION_TITLE_TEXT_LIMIT = 96
 const SESSION_PREVIEW_TEXT_LIMIT = 220
 const ELLIPSIS = '...'
@@ -42,15 +46,6 @@ export function normalizePreviewText(value: string): string | null {
   return finalizeNormalizedText(normalizeStringText(value, SESSION_PREVIEW_TEXT_LIMIT))
 }
 
-/** Cut to `limit` UTF-16 code units without splitting a trailing surrogate pair. */
-export function sliceAtCodeUnitLimit(value: string, limit: number): string {
-  if (value.length <= limit) {
-    return value
-  }
-  const end = limit > 0 && isHighSurrogate(value.charCodeAt(limit - 1)) ? limit - 1 : limit
-  return value.slice(0, end)
-}
-
 function normalizeContentText(value: unknown, limit: number): string | null {
   if (typeof value === 'string') {
     return finalizeNormalizedText(normalizeStringText(value, limit))
diff --git a/src/shared/surrogate-safe-text-slice.ts b/src/shared/surrogate-safe-text-slice.ts
new file mode 100644
index 00000000000..6e9e2d8d030
--- /dev/null
+++ b/src/shared/surrogate-safe-text-slice.ts
@@ -0,0 +1,19 @@
+// Cutting text to a length bound without splitting a character in half.
+//
+// JavaScript string length counts UTF-16 code units, so a raw `slice` at a
+// bound can land between the two halves of an astral character — an emoji, or
+// most CJK extension characters — and leave a lone surrogate that every surface
+// renders as U+FFFD.
+
+/** Cut to `limit` UTF-16 code units without splitting a trailing surrogate pair. */
+export function sliceAtCodeUnitLimit(value: string, limit: number): string {
+  if (value.length <= limit) {
+    return value
+  }
+  const end = limit > 0 && isHighSurrogate(value.charCodeAt(limit - 1)) ? limit - 1 : limit
+  return value.slice(0, end)
+}
+
+function isHighSurrogate(code: number): boolean {
+  return code >= 0xd800 && code <= 0xdbff
+}