AK: Introduce StringBase::replace_with_new_{short_,}string

Author: https://github.com/DanShaders
Commit: https://github.com/SerenityOS/serenity/commit/dcd1fda9c8
Pull-request: https://github.com/SerenityOS/serenity/pull/21661
Reviewed-by: https://github.com/ADKaster
Reviewed-by: https://github.com/alimpfard
Reviewed-by: https://github.com/kleinesfilmroellchen
Reviewed-by: https://github.com/trflynn89

AK/String.cpp

@@ -64,18 +64,6 @@ ErrorOr<NonnullRefPtr<StringData>> StringData::create_uninitialized(size_t byte_
     return new_string_data;
 }
 
-ErrorOr<NonnullRefPtr<StringData>> StringData::from_utf8(char const* utf8_data, size_t byte_count)
-{
-    // Strings of MAX_SHORT_STRING_BYTE_COUNT bytes or less should be handled by the String short string optimization.
-    VERIFY(byte_count > String::MAX_SHORT_STRING_BYTE_COUNT);
-
-    VERIFY(utf8_data);
-    u8* buffer = nullptr;
-    auto new_string_data = TRY(create_uninitialized(byte_count, buffer));
-    memcpy(buffer, utf8_data, byte_count * sizeof(char));
-    return new_string_data;
-}
-
 static ErrorOr<void> read_stream_into_buffer(Stream& stream, Bytes buffer)
 {
     TRY(stream.read_until_filled(buffer));
@@ -124,15 +112,12 @@ void StringData::compute_hash() const
 
 String String::from_utf8_without_validation(ReadonlyBytes bytes)
 {
-    if (bytes.size() <= MAX_SHORT_STRING_BYTE_COUNT) {
+    String result;
-        ShortString short_string;
+    MUST(result.replace_with_new_string(bytes.size(), [&](Bytes buffer) {
-        if (!bytes.is_empty())
+        bytes.copy_to(buffer);
-            memcpy(short_string.storage, bytes.data(), bytes.size());
+        return ErrorOr<void> {};
-        short_string.byte_count_and_short_string_flag = (bytes.size() << 1) | SHORT_STRING_FLAG;
+    }));
-        return String { short_string };
+    return result;
-    }
-    auto data = MUST(Detail::StringData::from_utf8(reinterpret_cast<char const*>(bytes.data()), bytes.size()));
-    return String { move(data) };
 }
 
 ErrorOr<String> String::from_utf8(StringView view)
@@ -140,15 +125,12 @@ ErrorOr<String> String::from_utf8(StringView view)
     if (!Utf8View { view }.validate())
         return Error::from_string_literal("String::from_utf8: Input was not valid UTF-8");
 
-    if (view.length() <= MAX_SHORT_STRING_BYTE_COUNT) {
+    String result;
-        ShortString short_string;
+    TRY(result.replace_with_new_string(view.length(), [&](Bytes buffer) {
-        if (!view.is_empty())
+        view.bytes().copy_to(buffer);
-            memcpy(short_string.storage, view.characters_without_null_termination(), view.length());
+        return ErrorOr<void> {};
-        short_string.byte_count_and_short_string_flag = (view.length() << 1) | SHORT_STRING_FLAG;
+    }));
-        return String { short_string };
+    return result;
-    }
-    auto data = TRY(Detail::StringData::from_utf8(view.characters_without_null_termination(), view.length()));
-    return String { move(data) };
 }
 
 ErrorOr<String> String::from_stream(Stream& stream, size_t byte_count)

AK/StringBase.cpp

@@ -54,11 +54,6 @@ StringBase& StringBase::operator=(StringBase const& other)
     return *this;
 }
 
-bool StringBase::is_short_string() const
-{
-    return has_short_string_bit(reinterpret_cast<uintptr_t>(m_data));
-}
-
 ReadonlyBytes StringBase::bytes() const
 {
     if (is_short_string())
@@ -82,6 +77,17 @@ bool StringBase::operator==(StringBase const& other) const
     return bytes() == other.bytes();
 }
 
+ErrorOr<Bytes> StringBase::replace_with_uninitialized_buffer(size_t byte_count)
+{
+    if (byte_count <= MAX_SHORT_STRING_BYTE_COUNT)
+        return replace_with_uninitialized_short_string(byte_count);
+
+    u8* buffer = nullptr;
+    destroy_string();
+    m_data = &TRY(StringData::create_uninitialized(byte_count, buffer)).leak_ref();
+    return Bytes { buffer, byte_count };
+}
+
 void StringBase::destroy_string()
 {
     if (!is_short_string())

AK/StringBase.h

@@ -55,7 +55,10 @@ public:
     }
 
     // NOTE: This is primarily interesting to unit tests.
-    [[nodiscard]] bool is_short_string() const;
+    [[nodiscard]] constexpr bool is_short_string() const
+    {
+        return (m_short_string.byte_count_and_short_string_flag & SHORT_STRING_FLAG) != 0;
+    }
 
     // Returns the underlying UTF-8 encoded bytes.
     // NOTE: There is no guarantee about null-termination.
@@ -80,12 +83,41 @@ protected:
     {
     }
 
+    template<typename Func>
+    ErrorOr<void> replace_with_new_string(size_t byte_count, Func&& callback)
+    {
+        Bytes buffer = TRY(replace_with_uninitialized_buffer(byte_count));
+        if (byte_count != 0)
+            TRY(callback(buffer));
+        return {};
+    }
+
+    template<typename Func>
+    constexpr void replace_with_new_short_string(size_t byte_count, Func&& callback)
+    {
+        Bytes buffer = replace_with_uninitialized_short_string(byte_count);
+        if (byte_count != 0)
+            callback(buffer);
+    }
+
     union {
         ShortString m_short_string;
         Detail::StringData const* m_data { nullptr };
     };
 
 private:
+    ErrorOr<Bytes> replace_with_uninitialized_buffer(size_t byte_count);
+
+    constexpr Bytes replace_with_uninitialized_short_string(size_t byte_count)
+    {
+        VERIFY(is_short_string());
+        VERIFY(byte_count <= MAX_SHORT_STRING_BYTE_COUNT);
+
+        m_short_string = ShortString {};
+        m_short_string.byte_count_and_short_string_flag = (byte_count << 1) | SHORT_STRING_FLAG;
+        return { m_short_string.storage, byte_count };
+    }
+
     void destroy_string();
 };
 

AK/StringInternals.h

@@ -17,7 +17,6 @@ class StringData final : public RefCounted<StringData> {
 public:
     static ErrorOr<NonnullRefPtr<StringData>> create_uninitialized(size_t, u8*& buffer);
     static ErrorOr<NonnullRefPtr<StringData>> create_substring(StringData const& superstring, size_t start, size_t byte_count);
-    static ErrorOr<NonnullRefPtr<StringData>> from_utf8(char const* utf8_bytes, size_t);
     static ErrorOr<NonnullRefPtr<StringData>> from_stream(Stream&, size_t byte_count);
 
     struct SubstringData {