gre-ilya/libuv
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

idna: fix assert in wtf8_to_utf16 conversion

Browse Source 
0x10FFFF is the valid max unicode character, so the check should be
inclusive.

This assert gets triggered because uv__wtf8_decode1 (used by
uv_wtf8_length_as_utf16) has the correct code_point <= 0x10FFFF check,
so the sequence is treated as valid and it will be passed into
uv_wtf8_to_utf16, where the incorrect assert gets triggered:

src/idna.c:397: uv_wtf8_to_utf16: Assertion `code_point < 0x10FFFF' failed.
This commit is contained in:
Tobiasz Laskowski 2026-02-03 23:04:36 +00:00
parent fc90bb9843
commit 526b323e51
2 changed files with 9 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/idna.c
View File

@ -394,7 +394,7 @@ void uv_wtf8_to_utf16(const char* source_ptr,
/* uv_wtf8_length_as_utf16 should have been called and checked first. */ /* uv_wtf8_length_as_utf16 should have been called and checked first. */
assert(code_point >= 0); assert(code_point >= 0);
if (code_point > 0xFFFF) { if (code_point > 0xFFFF) {
assert(code_point < 0x10FFFF); assert(code_point <= 0x10FFFF);
*w_target++ = (((code_point - 0x10000) >> 10) + 0xD800); *w_target++ = (((code_point - 0x10000) >> 10) + 0xD800);
*w_target++ = ((code_point - 0x10000) & 0x3FF) + 0xDC00; *w_target++ = ((code_point - 0x10000) & 0x3FF) + 0xDC00;
w_target_len -= 2; w_target_len -= 2;

8
test/test-idna.c
View File

@ -236,5 +236,13 @@ TEST_IMPL(wtf8) {
ASSERT_GT(len, 0); ASSERT_GT(len, 0);
ASSERT_LT(len, ARRAY_SIZE(buf)); ASSERT_LT(len, ARRAY_SIZE(buf));
uv_wtf8_to_utf16(input, buf, len); uv_wtf8_to_utf16(input, buf, len);
/* Test 0x10FFFF, max unicode character */
static const char input_max[] = "\xF4\x8F\xBF\xBF";
len = uv_wtf8_length_as_utf16(input_max);
ASSERT_GT(len, 0);
ASSERT_LT(len, ARRAY_SIZE(buf));
uv_wtf8_to_utf16(input_max, buf, len);
return 0; return 0;
} }