fix(chunk): keep emoji whole when length-splitting a long line

ly-wang19 · claude · ly-wang19 · commit 259f022f42d9 · 2026-06-24T20:39:42.000+08:00
chunkByNewline length-splits a long line with a raw
`lineValue.slice(0, firstLimit)`, which can cut through a surrogate
pair when firstLimit lands on an odd UTF-16 offset. The first chunk
then ends on a lone high surrogate and the next chunk begins with the
orphaned low surrogate, rendering a broken/replacement glyph.

Guard the boundary with avoidTrailingHighSurrogateBreak (already used by
chunkText/chunkMarkdownText in this file and already imported), so the
split lands on a code-point boundary. Only the rare mid-surrogate
boundary changes; all other inputs are byte-identical.

Co-Authored-By: Claude Opus 4.8 (1M context) &lt;noreply@anthropic.com&gt;
diff --git a/src/auto-reply/chunk.test.ts b/src/auto-reply/chunk.test.ts
@@ -513,6 +513,22 @@ describe("chunkByNewline", () => {
   it.each(["", "   \n\n   "] as const)("returns empty array for input %j", (text) => {
     expect(chunkByNewline(text, 100)).toStrictEqual([]);
   });
+
+  it("does not split surrogate pairs when length-splitting a long line", () => {
+    // A single line of 8 emoji (16 UTF-16 units) with an odd maxLineLength that
+    // would otherwise land mid-surrogate-pair at index 3.
+    const text = "😀".repeat(8);
+    const chunks = chunkByNewline(text, 3);
+
+    expect(chunks.join("")).toBe(text);
+    expect(chunks.length).toBeGreaterThan(1);
+    // No chunk ends on a lone high surrogate.
+    expect(chunks.every((chunk) => !/[\uD800-\uDBFF]$/u.test(chunk))).toBe(true);
+    // No chunk starts on a lone low surrogate.
+    expect(chunks.every((chunk) => !/^[\uDC00-\uDFFF]/u.test(chunk))).toBe(true);
+    // First chunk carries a single whole emoji rather than 😀 + lone surrogate.
+    expect(chunks[0]).toBe("😀");
+  });
 });
 
 describe("chunkTextWithMode", () => {
diff --git a/src/auto-reply/chunk.ts b/src/auto-reply/chunk.ts
@@ -164,9 +164,10 @@ export function chunkByNewline(
     }
 
     const firstLimit = Math.max(1, maxLineLength - prefix.length);
-    const first = lineValue.slice(0, firstLimit);
+    const safeLimit = avoidTrailingHighSurrogateBreak(lineValue, 0, firstLimit);
+    const first = lineValue.slice(0, safeLimit);
     chunks.push(prefix + first);
-    const remaining = lineValue.slice(firstLimit);
+    const remaining = lineValue.slice(safeLimit);
     if (remaining) {
       chunks.push(...chunkText(remaining, maxLineLength));
     }

Original file line number	Diff line number	Diff line change
`@@ -164,9 +164,10 @@ export function chunkByNewline(`
`164`	`164`	`}`
`165`	`165`
`166`	`166`	`const firstLimit = Math.max(1, maxLineLength - prefix.length);`
`167`		`- const first = lineValue.slice(0, firstLimit);`
	`167`	`+ const safeLimit = avoidTrailingHighSurrogateBreak(lineValue, 0, firstLimit);`
	`168`	`+ const first = lineValue.slice(0, safeLimit);`
`168`	`169`	`chunks.push(prefix + first);`
`169`		`- const remaining = lineValue.slice(firstLimit);`
	`170`	`+ const remaining = lineValue.slice(safeLimit);`
`170`	`171`	`if (remaining) {`
`171`	`172`	`chunks.push(...chunkText(remaining, maxLineLength));`
`172`	`173`	`}`