mirrors/ladybird
1
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2025-01-22 09:12:13 -05:00
Code Issues Projects Releases Packages Wiki Activity

LibUnicode: Consistently reject out-of-bounds segmenter indices

Browse source 
In the UTF-8 implementation, this prevents out-of-bounds access of the
underlying text data, as the ICU macro would essentially do something
akin to `text[text.length()]`.

The UTF-16 implementation already checks for out-of-bounds, but would
previously return 0. We now return an empty Optional in both impls. This
doesn't affect LibJS (the user of the UTF-16 impl), as it already does
bounds checking before invoking LibUnicode APIs.
This commit is contained in:
Timothy Flynn 2025-01-16 12:50:40 -05:00 committed by Jelle Raaijmakers
parent 0c09a099a5
commit e6b7c8cde2
Notes: github-actions[bot] 2025-01-16 22:23:49 +00:00 
Author: https://github.com/trflynn89
Commit: https://github.com/LadybirdBrowser/ladybird/commit/e6b7c8cde2b
Pull-request: https://github.com/LadybirdBrowser/ladybird/pull/3278
Reviewed-by: https://github.com/gmta ✅
2 changed files with 49 additions and 10 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

30
Libraries/LibUnicode/Segmenter.cpp
View file

@ -87,13 +87,15 @@ public:
virtual Optional<size_t> previous_boundary(size_t boundary, Inclusive inclusive) override
{
auto icu_boundary = align_boundary(boundary);
if (!icu_boundary.has_value())
return {};
if (inclusive == Inclusive::Yes) {
if (static_cast<bool>(m_segmenter->isBoundary(icu_boundary)))
return static_cast<size_t>(icu_boundary);
if (static_cast<bool>(m_segmenter->isBoundary(*icu_boundary)))
return static_cast<size_t>(*icu_boundary);
}
if (auto index = m_segmenter->preceding(icu_boundary); index != icu::BreakIterator::DONE)
if (auto index = m_segmenter->preceding(*icu_boundary); index != icu::BreakIterator::DONE)
return static_cast<size_t>(index);
return {};
@ -102,13 +104,15 @@ public:
virtual Optional<size_t> next_boundary(size_t boundary, Inclusive inclusive) override
{
auto icu_boundary = align_boundary(boundary);
if (!icu_boundary.has_value())
return {};
if (inclusive == Inclusive::Yes) {
if (static_cast<bool>(m_segmenter->isBoundary(icu_boundary)))
return static_cast<size_t>(icu_boundary);
if (static_cast<bool>(m_segmenter->isBoundary(*icu_boundary)))
return static_cast<size_t>(*icu_boundary);
}
if (auto index = m_segmenter->following(icu_boundary); index != icu::BreakIterator::DONE)
if (auto index = m_segmenter->following(*icu_boundary); index != icu::BreakIterator::DONE)
return static_cast<size_t>(index);
return {};
@ -173,19 +177,25 @@ public:
}
private:
i32 align_boundary(size_t boundary)
Optional<i32> align_boundary(size_t boundary)
{
auto icu_boundary = static_cast<i32>(boundary);
return m_segmented_text.visit(
[&](String const& text) {
[&](String const& text) -> Optional<i32> {
if (boundary >= text.byte_count())
return {};
U8_SET_CP_START(text.bytes().data(), 0, icu_boundary);
return icu_boundary;
},
[&](icu::UnicodeString const& text) {
[&](icu::UnicodeString const& text) -> Optional<i32> {
if (icu_boundary >= text.length())
return {};
return text.getChar32Start(icu_boundary);
},
[](Empty) -> i32 { VERIFY_NOT_REACHED(); });
[](Empty) -> Optional<i32> { VERIFY_NOT_REACHED(); });
}
void for_each_boundary(SegmentationCallback callback)

29
Tests/LibUnicode/TestSegmenter.cpp
View file

@ -9,6 +9,7 @@
#include <AK/Array.h>
#include <AK/String.h>
#include <AK/StringView.h>
#include <AK/Utf16View.h>
#include <AK/Vector.h>
#include <LibUnicode/Segmenter.h>
@ -126,3 +127,31 @@ TEST_CASE(word_segmentation)
"The quick (“brown”) fox can’t jump 32.3 feet, right?"sv,
{ 0u, 3u, 4u, 9u, 10u, 11u, 14u, 19u, 22u, 23u, 24u, 27u, 28u, 35u, 36u, 40u, 41u, 45u, 46u, 50u, 51u, 52u, 57u, 58u });
}
TEST_CASE(out_of_bounds)
{
{
auto text = "foo"_string;
auto segmenter = Unicode::Segmenter::create(Unicode::SegmenterGranularity::Word);
segmenter->set_segmented_text(text);
auto result = segmenter->previous_boundary(text.byte_count());
EXPECT(!result.has_value());
result = segmenter->next_boundary(text.byte_count());
EXPECT(!result.has_value());
}
{
auto text = MUST(AK::utf8_to_utf16("foo"sv));
auto segmenter = Unicode::Segmenter::create(Unicode::SegmenterGranularity::Word);
segmenter->set_segmented_text(Utf16View { text });
auto result = segmenter->previous_boundary(text.size());
EXPECT(!result.has_value());
result = segmenter->next_boundary(text.size());
EXPECT(!result.has_value());
}
}