Fix UTF-16-safe media runner output truncation

Alix-007 · Alix-007 · commit c34288039036 · 2026-07-07T11:02:48.000+08:00
diff --git a/src/media-understanding/runner.entries.ts b/src/media-understanding/runner.entries.ts
@@ -7,6 +7,7 @@ import {
   normalizeNullableString,
 } from "@openclaw/normalization-core/string-coerce";
 import { normalizeStringEntries } from "@openclaw/normalization-core/string-normalization";
+import { truncateUtf16Safe } from "@openclaw/normalization-core/utf16-slice";
 import { MediaUnderstandingSkipError } from "../../packages/media-understanding-common/src/errors.js";
 import { extractGeminiResponse } from "../../packages/media-understanding-common/src/output-extract.js";
 import {
@@ -100,7 +101,7 @@ function trimOutput(text: string, maxChars?: number): string {
   if (!maxChars || trimmed.length <= maxChars) {
     return trimmed;
   }
-  return trimmed.slice(0, maxChars).trim();
+  return truncateUtf16Safe(trimmed, maxChars).trim();
 }
 
 function extractSherpaOnnxText(raw: string): { matched: boolean; text: string } {
diff --git a/src/media-understanding/runner.video.test.ts b/src/media-understanding/runner.video.test.ts
@@ -41,6 +41,53 @@ function requireCapabilityOutput(result: CapabilityResult, index: number) {
 }
 
 describe("runCapability video provider wiring", () => {
+  it("truncates provider output without splitting a boundary emoji", async () => {
+    await withVideoFixture("openclaw-video-utf16-output", async ({ ctx, media, cache }) => {
+      const prefix = "v".repeat(79);
+      const result = await runCapability({
+        capability: "video",
+        cfg: {
+          models: {
+            providers: {
+              moonshot: {
+                apiKey: "test-key",
+                models: [],
+              },
+            },
+          },
+          tools: {
+            media: {
+              video: {
+                enabled: true,
+                models: [{ provider: "moonshot", model: "kimi-k2.5", maxChars: 80 }],
+              },
+            },
+          },
+        } as unknown as OpenClawConfig,
+        ctx,
+        attachments: cache,
+        media,
+        providerRegistry: new Map<string, MediaUnderstandingProvider>([
+          [
+            "moonshot",
+            {
+              id: "moonshot",
+              capabilities: ["video"],
+              describeVideo: async (req) => ({
+                text: `${prefix}${String.fromCodePoint(0x1f600)}tail`,
+                model: req.model,
+              }),
+            },
+          ],
+        ]),
+      });
+
+      const output = requireCapabilityOutput(result, 0);
+      expect(output.text).toBe(prefix);
+      expect(output.text).not.toContain(String.fromCharCode(0xd83d));
+    });
+  });
+
   it("merges video baseUrl and headers with entry precedence", async () => {
     let seenBaseUrl: string | undefined;
     let seenHeaders: Record<string, string> | undefined;
