LibJS: Port Intl.Segmenter to the ICU text segmenter

This also lets us fully implement detecting if a segment is word-like,
although that is not tested by test262.
This commit is contained in:
Timothy Flynn 2024-06-18 18:51:06 -04:00 committed by Andreas Kling
commit 14071c52f9
Notes: sideshowbarker 2024-07-17 03:45:48 +09:00
12 changed files with 95 additions and 162 deletions

View file

@ -13,7 +13,7 @@ namespace JS::Intl {
JS_DEFINE_ALLOCATOR(SegmentIterator);
// 18.6.1 CreateSegmentIterator ( segmenter, string ), https://tc39.es/ecma402/#sec-createsegmentsobject
NonnullGCPtr<SegmentIterator> SegmentIterator::create(Realm& realm, Segmenter& segmenter, Utf16View const& string, Segments const& segments)
NonnullGCPtr<SegmentIterator> SegmentIterator::create(Realm& realm, ::Locale::Segmenter const& segmenter, Utf16View const& string, Segments const& segments)
{
// 1. Let internalSlotsList be « [[IteratingSegmenter]], [[IteratedString]], [[IteratedStringNextSegmentCodeUnitIndex]] ».
// 2. Let iterator be OrdinaryObjectCreate(%SegmentIteratorPrototype%, internalSlotsList).
@ -21,22 +21,22 @@ NonnullGCPtr<SegmentIterator> SegmentIterator::create(Realm& realm, Segmenter& s
// 4. Set iterator.[[IteratedString]] to string.
// 5. Set iterator.[[IteratedStringNextSegmentCodeUnitIndex]] to 0.
// 6. Return iterator.
return realm.heap().allocate<SegmentIterator>(realm, realm, segmenter, move(string), segments);
return realm.heap().allocate<SegmentIterator>(realm, realm, segmenter, string, segments);
}
// 18.6 Segment Iterator Objects, https://tc39.es/ecma402/#sec-segment-iterator-objects
SegmentIterator::SegmentIterator(Realm& realm, Segmenter& segmenter, Utf16View const& string, Segments const& segments)
SegmentIterator::SegmentIterator(Realm& realm, ::Locale::Segmenter const& segmenter, Utf16View const& string, Segments const& segments)
: Object(ConstructWithPrototypeTag::Tag, realm.intrinsics().intl_segment_iterator_prototype())
, m_iterating_segmenter(segmenter)
, m_iterating_segmenter(segmenter.clone())
, m_iterated_string(string)
, m_segments(segments)
{
m_iterating_segmenter->set_segmented_text(m_iterated_string);
}
void SegmentIterator::visit_edges(Cell::Visitor& visitor)
{
Base::visit_edges(visitor);
visitor.visit(m_iterating_segmenter);
visitor.visit(m_segments);
}