RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-31 04:08:11 +00:00
Code Issues Activity Actions

AK+LibUnicode: Provide Unicode-aware caseless String matching

Browse source 
The Unicode spec defines much more complicated caseless matching
algorithms in its Collation spec. This implements the "basic" case
folding comparison.
This commit is contained in:
Timothy Flynn 2023-01-17 11:30:10 -05:00 • committed by Linus Groh
parent 8f2589b3b0
commit 537fcaf59e
3 changed files with 79 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

4
AK/String.h
View file

@ -49,6 +49,10 @@ public:
ErrorOr<String> to_lowercase(Optional<StringView> const& locale = {}) const; ErrorOr<String> to_lowercase(Optional<StringView> const& locale = {}) const;
ErrorOr<String> to_uppercase(Optional<StringView> const& locale = {}) const; ErrorOr<String> to_uppercase(Optional<StringView> const& locale = {}) const;
ErrorOr<String> to_titlecase(Optional<StringView> const& locale = {}) const; ErrorOr<String> to_titlecase(Optional<StringView> const& locale = {}) const;
ErrorOr<String> to_casefold() const;
// Compare this String against another string with caseless matching. Using this method requires linking LibUnicode into your application.
ErrorOr<bool> equals_ignoring_case(String const&) const;
// Creates a substring with a deep copy of the specified data window. // Creates a substring with a deep copy of the specified data window.
ErrorOr<String> substring_from_byte_offset(size_t start, size_t byte_count) const; ErrorOr<String> substring_from_byte_offset(size_t start, size_t byte_count) const;

60
Tests/AK/TestString.cpp
View file

@ -187,6 +187,66 @@ TEST_CASE(to_titlecase)
} }
} }
TEST_CASE(equals_ignoring_case)
{
{
String string1 {};
String string2 {};
EXPECT(MUST(string1.equals_ignoring_case(string2)));
}
{
auto string1 = MUST(String::from_utf8("abcd"sv));
auto string2 = MUST(String::from_utf8("ABCD"sv));
auto string3 = MUST(String::from_utf8("AbCd"sv));
auto string4 = MUST(String::from_utf8("dcba"sv));
EXPECT(MUST(string1.equals_ignoring_case(string2)));
EXPECT(MUST(string1.equals_ignoring_case(string3)));
EXPECT(!MUST(string1.equals_ignoring_case(string4)));
EXPECT(MUST(string2.equals_ignoring_case(string1)));
EXPECT(MUST(string2.equals_ignoring_case(string3)));
EXPECT(!MUST(string2.equals_ignoring_case(string4)));
EXPECT(MUST(string3.equals_ignoring_case(string1)));
EXPECT(MUST(string3.equals_ignoring_case(string2)));
EXPECT(!MUST(string3.equals_ignoring_case(string4)));
}
{
auto string1 = MUST(String::from_utf8("\u00DF"sv)); // LATIN SMALL LETTER SHARP S
auto string2 = MUST(String::from_utf8("SS"sv));
auto string3 = MUST(String::from_utf8("Ss"sv));
auto string4 = MUST(String::from_utf8("ss"sv));
auto string5 = MUST(String::from_utf8("S"sv));
auto string6 = MUST(String::from_utf8("s"sv));
EXPECT(MUST(string1.equals_ignoring_case(string2)));
EXPECT(MUST(string1.equals_ignoring_case(string3)));
EXPECT(MUST(string1.equals_ignoring_case(string4)));
EXPECT(!MUST(string1.equals_ignoring_case(string5)));
EXPECT(!MUST(string1.equals_ignoring_case(string6)));
EXPECT(MUST(string2.equals_ignoring_case(string1)));
EXPECT(MUST(string2.equals_ignoring_case(string3)));
EXPECT(MUST(string2.equals_ignoring_case(string4)));
EXPECT(!MUST(string2.equals_ignoring_case(string5)));
EXPECT(!MUST(string2.equals_ignoring_case(string6)));
EXPECT(MUST(string3.equals_ignoring_case(string1)));
EXPECT(MUST(string3.equals_ignoring_case(string2)));
EXPECT(MUST(string3.equals_ignoring_case(string4)));
EXPECT(!MUST(string3.equals_ignoring_case(string5)));
EXPECT(!MUST(string3.equals_ignoring_case(string6)));
EXPECT(MUST(string4.equals_ignoring_case(string1)));
EXPECT(MUST(string4.equals_ignoring_case(string2)));
EXPECT(MUST(string4.equals_ignoring_case(string3)));
EXPECT(!MUST(string4.equals_ignoring_case(string5)));
EXPECT(!MUST(string4.equals_ignoring_case(string6)));
}
}
TEST_CASE(is_one_of) TEST_CASE(is_one_of)
{ {
auto foo = MUST(String::from_utf8("foo"sv)); auto foo = MUST(String::from_utf8("foo"sv));

15
Userland/Libraries/LibUnicode/String.cpp
View file

@ -33,4 +33,19 @@ ErrorOr<String> String::to_titlecase(Optional<StringView> const& locale) const
return builder.to_string(); return builder.to_string();
} }
ErrorOr<String> String::to_casefold() const
{
StringBuilder builder;
TRY(Unicode::Detail::build_casefold_string(code_points(), builder));
return builder.to_string();
}
// https://www.unicode.org/versions/Unicode15.0.0/ch03.pdf#G34145
ErrorOr<bool> String::equals_ignoring_case(String const& other) const
{
// A string X is a caseless match for a string Y if and only if:
// toCasefold(X) = toCasefold(Y)
return TRY(to_casefold()) == TRY(other.to_casefold());
}
} }