From c6b67af39011e040267bff20deaeb254feb869ad Mon Sep 17 00:00:00 2001
From: Ben Noordhuis <info@bnoordhuis.nl>
Date: Tue, 19 Nov 2024 19:09:03 +0100
Subject: [PATCH] unix,win: fix off-by-one in uv_wtf8_to_utf16() (#4609)

uv_wtf8_length_as_utf16() checks if codepoints are > 0xFFFF (to see if
it should be encoded as a surrogate pair), therefore uv_wtf8_to_utf16()
should too. Instead it checked > 0x1000. Harmonize the checks.

Fixes: https://github.com/nodejs/node/issues/55914
---
 src/idna.c       |  2 +-
 test/test-idna.c | 12 ++++++++++++
 test/test-list.h |  2 ++
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/src/idna.c b/src/idna.c
index efc5f283..5fcaf64c 100644
--- a/src/idna.c
+++ b/src/idna.c
@@ -393,7 +393,7 @@ void uv_wtf8_to_utf16(const char* source_ptr,
     code_point = uv__wtf8_decode1(&source_ptr);
     /* uv_wtf8_length_as_utf16 should have been called and checked first. */
     assert(code_point >= 0);
-    if (code_point > 0x10000) {
+    if (code_point > 0xFFFF) {
       assert(code_point < 0x10FFFF);
       *w_target++ = (((code_point - 0x10000) >> 10) + 0xD800);
       *w_target++ = ((code_point - 0x10000) & 0x3FF) + 0xDC00;
diff --git a/test/test-idna.c b/test/test-idna.c
index 28f9eaaa..4a165573 100644
--- a/test/test-idna.c
+++ b/test/test-idna.c
@@ -218,3 +218,15 @@ TEST_IMPL(idna_toascii) {
 #undef T
 
 #endif  /* __MVS__ */
+
+TEST_IMPL(wtf8) {
+  static const char input[] = "ᜄȺy𐞲:𞢢𘴇𐀀'¥3̞[<i$";
+  uint16_t buf[32];
+  ssize_t len;
+
+  len = uv_wtf8_length_as_utf16(input);
+  ASSERT_GT(len, 0);
+  ASSERT_LT(len, ARRAY_SIZE(buf));
+  uv_wtf8_to_utf16(input, buf, len);
+  return 0;
+}
diff --git a/test/test-list.h b/test/test-list.h
index e07bd61e..a64c2c33 100644
--- a/test/test-list.h
+++ b/test/test-list.h
@@ -572,6 +572,7 @@ TEST_DECLARE  (fork_threadpool_queue_work_simple)
 
 TEST_DECLARE  (iouring_pollhup)
 
+TEST_DECLARE  (wtf8)
 TEST_DECLARE  (idna_toascii)
 TEST_DECLARE  (utf8_decode1)
 TEST_DECLARE  (utf8_decode1_overrun)
@@ -1223,6 +1224,7 @@ TASK_LIST_START
 
   TEST_ENTRY  (iouring_pollhup)
 
+  TEST_ENTRY  (wtf8)
   TEST_ENTRY  (utf8_decode1)
   TEST_ENTRY  (utf8_decode1_overrun)
   TEST_ENTRY  (uname)