AK: Add a method to create a String with a repeated code point

Author: https://github.com/trflynn89
Commit: https://github.com/SerenityOS/serenity/commit/427b82065c
Pull-request: https://github.com/SerenityOS/serenity/pull/17132
Reviewed-by: https://github.com/linusg

AK/String.cpp

@@ -4,6 +4,7 @@
  * SPDX-License-Identifier: BSD-2-Clause
  */
 
+#include <AK/Array.h>
 #include <AK/Checked.h>
 #include <AK/FlyString.h>
 #include <AK/Format.h>
@@ -221,6 +222,44 @@ ErrorOr<String> String::from_utf8(StringView view)
     return String { move(data) };
 }
 
+ErrorOr<String> String::repeated(u32 code_point, size_t count)
+{
+    VERIFY(is_unicode(code_point));
+
+    Array<u8, 4> code_point_as_utf8;
+    size_t i = 0;
+
+    size_t code_point_byte_length = UnicodeUtils::code_point_to_utf8(code_point, [&](auto byte) {
+        code_point_as_utf8[i++] = static_cast<u8>(byte);
+    });
+
+    auto copy_to_buffer = [&](u8* buffer) {
+        if (code_point_byte_length == 1) {
+            memset(buffer, code_point_as_utf8[0], count);
+            return;
+        }
+
+        for (i = 0; i < count; ++i)
+            memcpy(buffer + (i * code_point_byte_length), code_point_as_utf8.data(), code_point_byte_length);
+    };
+
+    auto total_byte_count = code_point_byte_length * count;
+
+    if (total_byte_count <= MAX_SHORT_STRING_BYTE_COUNT) {
+        ShortString short_string;
+        copy_to_buffer(short_string.storage);
+        short_string.byte_count_and_short_string_flag = (total_byte_count << 1) | SHORT_STRING_FLAG;
+
+        return String { short_string };
+    }
+
+    u8* buffer = nullptr;
+    auto new_string_data = TRY(Detail::StringData::create_uninitialized(total_byte_count, buffer));
+    copy_to_buffer(buffer);
+
+    return String { move(new_string_data) };
+}
+
 StringView String::bytes_as_string_view() const
 {
     return StringView(bytes());

AK/String.h

@@ -93,6 +93,9 @@ public:
         return String { short_string };
     }
 
+    // Creates a new String with a single code point repeated N times.
+    static ErrorOr<String> repeated(u32 code_point, size_t count);
+
     // Creates a new String by case-transforming this String. Using these methods require linking LibUnicode into your application.
     ErrorOr<String> to_lowercase(Optional<StringView> const& locale = {}) const;
     ErrorOr<String> to_uppercase(Optional<StringView> const& locale = {}) const;

Tests/AK/TestString.cpp

@@ -357,3 +357,81 @@ TEST_CASE(find_byte_offset)
         EXPECT_EQ(index5, 8u);
     }
 }
+
+TEST_CASE(repeated)
+{
+    {
+        auto string1 = MUST(String::repeated('a', 0));
+        EXPECT(string1.is_short_string());
+        EXPECT(string1.is_empty());
+
+        auto string2 = MUST(String::repeated(0x03C9U, 0));
+        EXPECT(string2.is_short_string());
+        EXPECT(string2.is_empty());
+
+        auto string3 = MUST(String::repeated(0x10300, 0));
+        EXPECT(string3.is_short_string());
+        EXPECT(string3.is_empty());
+    }
+    {
+        auto string1 = MUST(String::repeated('a', 1));
+        EXPECT(string1.is_short_string());
+        EXPECT_EQ(string1.bytes_as_string_view().length(), 1u);
+        EXPECT_EQ(string1, "a"sv);
+
+        auto string2 = MUST(String::repeated(0x03C9U, 1));
+        EXPECT(string2.is_short_string());
+        EXPECT_EQ(string2.bytes_as_string_view().length(), 2u);
+        EXPECT_EQ(string2, "ω"sv);
+
+        auto string3 = MUST(String::repeated(0x10300, 1));
+#ifdef AK_ARCH_64_BIT
+        EXPECT(string3.is_short_string());
+#else
+        EXPECT(!string3.is_short_string());
+#endif
+        EXPECT_EQ(string3.bytes_as_string_view().length(), 4u);
+        EXPECT_EQ(string3, "𐌀"sv);
+    }
+    {
+        auto string1 = MUST(String::repeated('a', 3));
+        EXPECT(string1.is_short_string());
+        EXPECT_EQ(string1.bytes_as_string_view().length(), 3u);
+        EXPECT_EQ(string1, "aaa"sv);
+
+        auto string2 = MUST(String::repeated(0x03C9U, 3));
+#ifdef AK_ARCH_64_BIT
+        EXPECT(string2.is_short_string());
+#else
+        EXPECT(!string2.is_short_string());
+#endif
+        EXPECT_EQ(string2.bytes_as_string_view().length(), 6u);
+        EXPECT_EQ(string2, "ωωω"sv);
+
+        auto string3 = MUST(String::repeated(0x10300, 3));
+        EXPECT(!string3.is_short_string());
+        EXPECT_EQ(string3.bytes_as_string_view().length(), 12u);
+        EXPECT_EQ(string3, "𐌀𐌀𐌀"sv);
+    }
+    {
+        auto string1 = MUST(String::repeated('a', 10));
+        EXPECT(!string1.is_short_string());
+        EXPECT_EQ(string1.bytes_as_string_view().length(), 10u);
+        EXPECT_EQ(string1, "aaaaaaaaaa"sv);
+
+        auto string2 = MUST(String::repeated(0x03C9U, 10));
+        EXPECT(!string2.is_short_string());
+        EXPECT_EQ(string2.bytes_as_string_view().length(), 20u);
+        EXPECT_EQ(string2, "ωωωωωωωωωω"sv);
+
+        auto string3 = MUST(String::repeated(0x10300, 10));
+        EXPECT(!string3.is_short_string());
+        EXPECT_EQ(string3.bytes_as_string_view().length(), 40u);
+        EXPECT_EQ(string3, "𐌀𐌀𐌀𐌀𐌀𐌀𐌀𐌀𐌀𐌀"sv);
+    }
+
+    EXPECT_CRASH("Creating a string from an invalid code point", [] {
+        (void)String::repeated(0xffffffff, 1);
+        return Test::Crash::Failure::DidNotCrash;
+    });
+}