AK: Add a String factory to create a string from a single code point

2 files changed, 43 insertions, 1 deletions

diff --git a/AK/String.h b/AK/String.h
index dc73fe7d55..7aca7b50dc 100644
--- a/AK/String.h
+++ b/AK/String.h
@@ -1,11 +1,13 @@
 /*
  * Copyright (c) 2018-2022, Andreas Kling <kling@serenityos.org>
+ * Copyright (c) 2023, Tim Flynn <trflynn89@serenityos.org>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
 
 #pragma once
 
+#include <AK/CharacterTypes.h>
 #include <AK/Concepts.h>
 #include <AK/Format.h>
 #include <AK/Forward.h>
@@ -16,6 +18,7 @@
 #include <AK/StringView.h>
 #include <AK/Traits.h>
 #include <AK/Types.h>
+#include <AK/UnicodeUtils.h>
 #include <AK/Vector.h>
 
 namespace AK {
@@ -74,6 +77,22 @@ public:
         return String { short_string };
     }
 
+    // Creates a new String from a single code point.
+    static constexpr String from_code_point(u32 code_point)
+    {
+        VERIFY(is_unicode(code_point));
+
+        ShortString short_string;
+        size_t i = 0;
+
+        auto length = UnicodeUtils::code_point_to_utf8(code_point, [&](auto byte) {
+            short_string.storage[i++] = static_cast<u8>(byte);
+        });
+        short_string.byte_count_and_short_string_flag = (length << 1) | SHORT_STRING_FLAG;
+
+        return String { short_string };
+    }
+
     // 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;
diff --git a/Tests/AK/TestString.cpp b/Tests/AK/TestString.cpp
index c5785c7dab..f90b7bf393 100644
--- a/Tests/AK/TestString.cpp
+++ b/Tests/AK/TestString.cpp
@@ -4,9 +4,12 @@
  * SPDX-License-Identifier: BSD-2-Clause
  */
 
+// This is included first on purpose. We specifically do not want LibTest to override VERIFY here so
+// that we can actually test that some String factory methods cause a crash with invalid input.
+#include <AK/String.h>
+
 #include <LibTest/TestCase.h>
 
-#include <AK/String.h>
 #include <AK/StringBuilder.h>
 #include <AK/Try.h>
 #include <AK/Utf8View.h>
@@ -64,6 +67,26 @@ TEST_CASE(long_strings)
     EXPECT_EQ(string.bytes_as_string_view(), "abcdefgh"sv);
 }
 
+TEST_CASE(from_code_points)
+{
+    for (u32 code_point = 0; code_point < 0x80; ++code_point) {
+        auto string = String::from_code_point(code_point);
+
+        auto ch = static_cast<char>(code_point);
+        StringView view { &ch, 1 };
+
+        EXPECT_EQ(string, view);
+    }
+
+    auto string = String::from_code_point(0x10ffff);
+    EXPECT_EQ(string, "\xF4\x8F\xBF\xBF"sv);
+
+    EXPECT_CRASH("Creating a string from an invalid code point", [] {
+        String::from_code_point(0xffffffff);
+        return Test::Crash::Failure::DidNotCrash;
+    });
+}
+
 TEST_CASE(substring)
 {
     auto superstring = MUST(String::from_utf8("Hello I am a long string"sv));