LibJS+LibUnicode: Port Intl.Collator to UTF-16 strings

Author: https://github.com/trflynn89
Commit: b2f053e783
Pull-request: https://github.com/LadybirdBrowser/ladybird/pull/5581

Libraries/LibJS/Runtime/Intl/CollatorCompareFunction.cpp

@@ -48,17 +48,17 @@ ThrowCompletionOr<Value> CollatorCompareFunction::call()
     // 4. If y is not provided, let y be undefined.
 
     // 5. Let X be ? ToString(x).
-    auto x = TRY(vm.argument(0).to_string(vm));
+    auto x = TRY(vm.argument(0).to_utf16_string(vm));
 
     // 6. Let Y be ? ToString(y).
-    auto y = TRY(vm.argument(1).to_string(vm));
+    auto y = TRY(vm.argument(1).to_utf16_string(vm));
 
     // 7. Return CompareStrings(collator, X, Y).
     return compare_strings(m_collator, x, y);
 }
 
 // 10.3.3.2 CompareStrings ( collator, x, y ), https://tc39.es/ecma402/#sec-collator-comparestrings
-int compare_strings(Collator const& collator, StringView x, StringView y)
+int compare_strings(Collator const& collator, Utf16View const& x, Utf16View const& y)
 {
     auto result = collator.collator().compare(x, y);
 

Libraries/LibJS/Runtime/Intl/CollatorCompareFunction.h

@@ -30,6 +30,6 @@ private:
     GC::Ref<Collator> m_collator; // [[Collator]]
 };
 
-int compare_strings(Collator const&, StringView x, StringView y);
+int compare_strings(Collator const&, Utf16View const& x, Utf16View const& y);
 
 }

Libraries/LibJS/Runtime/StringPrototype.cpp

@@ -538,10 +538,10 @@ JS_DEFINE_NATIVE_FUNCTION(StringPrototype::locale_compare)
     auto object = TRY(require_object_coercible(vm, vm.this_value()));
 
     // 2. Let S be ? ToString(O).
-    auto string = TRY(object.to_string(vm));
+    auto string = TRY(object.to_utf16_string(vm));
 
     // 3. Let thatValue be ? ToString(that).
-    auto that_value = TRY(vm.argument(0).to_string(vm));
+    auto that_value = TRY(vm.argument(0).to_utf16_string(vm));
 
     // 4. Let collator be ? Construct(%Collator%, « locales, options »).
     auto locales = vm.argument(1);

Libraries/LibUnicode/Collator.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2024, Tim Flynn <trflynn89@serenityos.org>
+ * Copyright (c) 2024-2025, Tim Flynn <trflynn89@ladybird.org>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
@@ -168,11 +168,14 @@ public:
     {
     }
 
-    virtual Collator::Order compare(StringView lhs, StringView rhs) const override
+    virtual Collator::Order compare(Utf16View const& lhs, Utf16View const& rhs) const override
     {
         UErrorCode status = U_ZERO_ERROR;
 
-        auto result = m_collator->compareUTF8(icu_string_piece(lhs), icu_string_piece(rhs), status);
+        auto lhs_it = icu_string_iterator(lhs);
+        auto rhs_it = icu_string_iterator(rhs);
+
+        auto result = m_collator->compare(lhs_it, rhs_it, status);
         VERIFY(icu_success(status));
 
         switch (result) {

Libraries/LibUnicode/Collator.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2024, Tim Flynn <trflynn89@serenityos.org>
+ * Copyright (c) 2024-2025, Tim Flynn <trflynn89@ladybird.org>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
@@ -53,7 +53,7 @@ public:
         Equal,
         After,
     };
-    virtual Order compare(StringView, StringView) const = 0;
+    virtual Order compare(Utf16View const&, Utf16View const&) const = 0;
 
     virtual Sensitivity sensitivity() const = 0;
     virtual bool ignore_punctuation() const = 0;

Libraries/LibUnicode/ICU.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2024, Tim Flynn <trflynn89@serenityos.org>
+ * Copyright (c) 2024-2025, Tim Flynn <trflynn89@ladybird.org>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
@@ -162,4 +162,16 @@ String icu_string_to_string(UChar const* string, i32 length)
     return MUST(Utf16View { string, static_cast<size_t>(length) }.to_utf8());
 }
 
+UCharIterator icu_string_iterator(Utf16View const& string)
+{
+    UCharIterator iterator;
+
+    if (string.has_ascii_storage())
+        uiter_setUTF8(&iterator, string.ascii_span().data(), static_cast<i32>(string.length_in_code_units()));
+    else
+        uiter_setString(&iterator, string.utf16_span().data(), static_cast<i32>(string.length_in_code_units()));
+
+    return iterator;
+}
+
 }

Libraries/LibUnicode/ICU.h

@@ -16,6 +16,7 @@
 #include <unicode/locid.h>
 #include <unicode/strenum.h>
 #include <unicode/stringpiece.h>
+#include <unicode/uiter.h>
 #include <unicode/uloc.h>
 #include <unicode/unistr.h>
 #include <unicode/utypes.h>
@@ -102,6 +103,8 @@ Vector<icu::UnicodeString> icu_string_list(ReadonlySpan<String> strings);
 String icu_string_to_string(icu::UnicodeString const& string);
 String icu_string_to_string(UChar const*, i32 length);
 
+UCharIterator icu_string_iterator(Utf16View const&);
+
 template<typename Filter>
 Vector<String> icu_string_enumeration_to_list(OwnPtr<icu::StringEnumeration> enumeration, char const* bcp47_keyword, Filter&& filter)
 {