From 3f04f3939ea0ac8fdd766a60655d29de2ffb44e5 Mon Sep 17 00:00:00 2001 From: Jonathan Wakely Date: Wed, 1 May 2024 17:09:39 +0100 Subject: [PATCH] libstdc++: Fix handling of incomplete UTF-8 sequences in _Unicode_view Eddie Nolan reported to me that _Unicode_view was not correctly implementing the substitution of ill-formed subsequences with U+FFFD, due to failing to increment the counter when the iterator reaches the end of the sequence before a multibyte sequence is complete. As a result, the incomplete sequence was not completely consumed, and then the remaining character was treated as another ill-formed sequence, giving two U+FFFD characters instead of one. To avoid similar mistakes in future, this change introduces a lambda that increments the iterator and the counter together. This ensures the counter is always incremented when the iterator is incremented, so that we always know how many characters have been consumed. libstdc++-v3/ChangeLog: * include/bits/unicode.h (_Unicode_view::_M_read_utf8): Ensure count of characters consumed is correct when the end of the input is reached unexpectedly. * testsuite/ext/unicode/view.cc: Test incomplete UTF-8 sequences. --- libstdc++-v3/include/bits/unicode.h | 24 +++++++++++------------- libstdc++-v3/testsuite/ext/unicode/view.cc | 7 +++++++ 2 files changed, 18 insertions(+), 13 deletions(-) diff --git a/libstdc++-v3/include/bits/unicode.h b/libstdc++-v3/include/bits/unicode.h index 29813b743dc..46238143fb6 100644 --- a/libstdc++-v3/include/bits/unicode.h +++ b/libstdc++-v3/include/bits/unicode.h @@ -261,9 +261,13 @@ namespace __unicode { _Guard<_Iter> __g{this, _M_curr()}; char32_t __c{}; - uint8_t __u = *_M_curr()++; const uint8_t __lo_bound = 0x80, __hi_bound = 0xBF; + uint8_t __u = *_M_curr()++; uint8_t __to_incr = 1; + auto __incr = [&, this] { + ++__to_incr; + return ++_M_curr(); + }; if (__u <= 0x7F) [[likely]] // 0x00 to 0x7F __c = __u; @@ -281,8 +285,7 @@ namespace __unicode else { __c = (__c << 6) | (__u & 0x3F); - ++_M_curr(); - ++__to_incr; + __incr(); } } else if (__u <= 0xEF) // 0xE0 to 0xEF @@ -295,11 +298,10 @@ namespace __unicode if (__u < __lo_bound_2 || __u > __hi_bound_2) [[unlikely]] __c = _S_error(); - else if (++_M_curr() == _M_last) [[unlikely]] + else if (__incr() == _M_last) [[unlikely]] __c = _S_error(); else { - ++__to_incr; __c = (__c << 6) | (__u & 0x3F); __u = *_M_curr(); @@ -308,8 +310,7 @@ namespace __unicode else { __c = (__c << 6) | (__u & 0x3F); - ++_M_curr(); - ++__to_incr; + __incr(); } } } @@ -323,21 +324,19 @@ namespace __unicode if (__u < __lo_bound_2 || __u > __hi_bound_2) [[unlikely]] __c = _S_error(); - else if (++_M_curr() == _M_last) [[unlikely]] + else if (__incr() == _M_last) [[unlikely]] __c = _S_error(); else { - ++__to_incr; __c = (__c << 6) | (__u & 0x3F); __u = *_M_curr(); if (__u < __lo_bound || __u > __hi_bound) [[unlikely]] __c = _S_error(); - else if (++_M_curr() == _M_last) [[unlikely]] + else if (__incr() == _M_last) [[unlikely]] __c = _S_error(); else { - ++__to_incr; __c = (__c << 6) | (__u & 0x3F); __u = *_M_curr(); @@ -346,8 +345,7 @@ namespace __unicode else { __c = (__c << 6) | (__u & 0x3F); - ++_M_curr(); - ++__to_incr; + __incr(); } } } diff --git a/libstdc++-v3/testsuite/ext/unicode/view.cc b/libstdc++-v3/testsuite/ext/unicode/view.cc index ee23b0b1d8a..6f3c099bd84 100644 --- a/libstdc++-v3/testsuite/ext/unicode/view.cc +++ b/libstdc++-v3/testsuite/ext/unicode/view.cc @@ -55,6 +55,13 @@ test_illformed_utf8() VERIFY( std::ranges::equal(v5, u8"\uFFFD\uFFFD\uFFFD\uFFFD\uFFFD\x41\uFFFD\uFFFD\x42"sv) ); uc::_Utf8_view v6("\xe1\x80\xe2\xf0\x91\x92\xf1\xbf\x41"sv); // Table 3-11 VERIFY( std::ranges::equal(v6, u8"\uFFFD\uFFFD\uFFFD\uFFFD\x41"sv) ); + + uc::_Utf32_view v7("\xe1\x80"sv); + VERIFY( std::ranges::equal(v7, U"\uFFFD"sv) ); + uc::_Utf32_view v8("\xf1\x80"sv); + VERIFY( std::ranges::equal(v8, U"\uFFFD"sv) ); + uc::_Utf32_view v9("\xf1\x80\x80"sv); + VERIFY( std::ranges::equal(v9, U"\uFFFD"sv) ); } constexpr void -- 2.11.4.GIT