LibTextCodec+Everywhere: Return Optional<Decoder&> from decoder_for()

Author: https://github.com/AtkinsSJ
Commit: https://github.com/SerenityOS/serenity/commit/f2a9426885
Pull-request: https://github.com/SerenityOS/serenity/pull/17511
Reviewed-by: https://github.com/ADKaster
Reviewed-by: https://github.com/nico

Meta/Lagom/Fuzzers/FuzzCyrillicDecoder.cpp

@@ -11,8 +11,8 @@
 
 extern "C" int LLVMFuzzerTestOneInput(uint8_t const* data, size_t size)
 {
-    auto* decoder = TextCodec::decoder_for("windows-1251"sv);
+    auto decoder = TextCodec::decoder_for("windows-1251"sv);
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
     decoder->to_utf8({ data, size });
     return 0;
 }

Meta/Lagom/Fuzzers/FuzzHebrewDecoder.cpp

@@ -11,8 +11,8 @@
 
 extern "C" int LLVMFuzzerTestOneInput(uint8_t const* data, size_t size)
 {
-    auto* decoder = TextCodec::decoder_for("windows-1255"sv);
+    auto decoder = TextCodec::decoder_for("windows-1255"sv);
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
     decoder->to_utf8({ data, size });
     return 0;
 }

Meta/Lagom/Fuzzers/FuzzLatin1Decoder.cpp

@@ -11,8 +11,8 @@
 
 extern "C" int LLVMFuzzerTestOneInput(uint8_t const* data, size_t size)
 {
-    auto* decoder = TextCodec::decoder_for("windows-1252"sv);
+    auto decoder = TextCodec::decoder_for("windows-1252"sv);
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
     decoder->to_utf8({ data, size });
     return 0;
 }

Meta/Lagom/Fuzzers/FuzzLatin2Decoder.cpp

@@ -11,8 +11,8 @@
 
 extern "C" int LLVMFuzzerTestOneInput(uint8_t const* data, size_t size)
 {
-    auto* decoder = TextCodec::decoder_for("iso-8859-2"sv);
+    auto decoder = TextCodec::decoder_for("iso-8859-2"sv);
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
     decoder->to_utf8({ data, size });
     return 0;
 }

Meta/Lagom/Fuzzers/FuzzUTF16BEDecoder.cpp

@@ -11,8 +11,8 @@
 
 extern "C" int LLVMFuzzerTestOneInput(uint8_t const* data, size_t size)
 {
-    auto* decoder = TextCodec::decoder_for("utf-16be"sv);
+    auto decoder = TextCodec::decoder_for("utf-16be"sv);
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
     decoder->to_utf8({ data, size });
     return 0;
 }

Userland/Libraries/LibTextCodec/Decoder.cpp

@@ -1,6 +1,7 @@
 /*
  * Copyright (c) 2020, Andreas Kling <kling@serenityos.org>
  * Copyright (c) 2022, Jelle Raaijmakers <jelle@gmta.nl>
+ * Copyright (c) 2023, Sam Atkins <atkinssj@serenityos.org>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
@@ -30,37 +31,37 @@ TurkishDecoder s_turkish_decoder;
 XUserDefinedDecoder s_x_user_defined_decoder;
 }
 
-Decoder* decoder_for(StringView a_encoding)
+Optional<Decoder&> decoder_for(StringView a_encoding)
 {
     auto encoding = get_standardized_encoding(a_encoding);
     if (encoding.has_value()) {
         if (encoding.value().equals_ignoring_case("windows-1252"sv))
-            return &s_latin1_decoder;
+            return s_latin1_decoder;
         if (encoding.value().equals_ignoring_case("utf-8"sv))
-            return &s_utf8_decoder;
+            return s_utf8_decoder;
         if (encoding.value().equals_ignoring_case("utf-16be"sv))
-            return &s_utf16be_decoder;
+            return s_utf16be_decoder;
         if (encoding.value().equals_ignoring_case("utf-16le"sv))
-            return &s_utf16le_decoder;
+            return s_utf16le_decoder;
         if (encoding.value().equals_ignoring_case("iso-8859-2"sv))
-            return &s_latin2_decoder;
+            return s_latin2_decoder;
         if (encoding.value().equals_ignoring_case("windows-1255"sv))
-            return &s_hebrew_decoder;
+            return s_hebrew_decoder;
         if (encoding.value().equals_ignoring_case("windows-1251"sv))
-            return &s_cyrillic_decoder;
+            return s_cyrillic_decoder;
         if (encoding.value().equals_ignoring_case("koi8-r"sv))
-            return &s_koi8r_decoder;
+            return s_koi8r_decoder;
         if (encoding.value().equals_ignoring_case("iso-8859-15"sv))
-            return &s_latin9_decoder;
+            return s_latin9_decoder;
         if (encoding.value().equals_ignoring_case("macintosh"sv))
-            return &s_mac_roman_decoder;
+            return s_mac_roman_decoder;
         if (encoding.value().equals_ignoring_case("windows-1254"sv))
-            return &s_turkish_decoder;
+            return s_turkish_decoder;
         if (encoding.value().equals_ignoring_case("x-user-defined"sv))
-            return &s_x_user_defined_decoder;
+            return s_x_user_defined_decoder;
     }
     dbgln("TextCodec: No decoder implemented for encoding '{}'", a_encoding);
-    return nullptr;
+    return {};
 }
 
 // https://encoding.spec.whatwg.org/#concept-encoding-get

Userland/Libraries/LibTextCodec/Decoder.h

@@ -1,6 +1,7 @@
 /*
  * Copyright (c) 2020-2021, Andreas Kling <kling@serenityos.org>
  * Copyright (c) 2022, Jelle Raaijmakers <jelle@gmta.nl>
+ * Copyright (c) 2023, Sam Atkins <atkinssj@serenityos.org>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
@@ -9,6 +10,7 @@
 
 #include <AK/Forward.h>
 #include <AK/Function.h>
+#include <AK/Optional.h>
 
 namespace TextCodec {
 
@@ -84,7 +86,7 @@ public:
     virtual void process(StringView, Function<void(u32)> on_code_point) override;
 };
 
-Decoder* decoder_for(StringView encoding);
+Optional<Decoder&> decoder_for(StringView encoding);
 Optional<StringView> get_standardized_encoding(StringView encoding);
 
 // This returns the appropriate Unicode decoder for the sniffed BOM or nullptr if there is no appropriate decoder.

Userland/Libraries/LibWeb/CSS/Parser/Tokenizer.cpp

@@ -199,8 +199,8 @@ Tokenizer::Tokenizer(StringView input, StringView encoding)
 {
     // https://www.w3.org/TR/css-syntax-3/#css-filter-code-points
     auto filter_code_points = [](StringView input, auto encoding) -> ErrorOr<String> {
-        auto* decoder = TextCodec::decoder_for(encoding);
+        auto decoder = TextCodec::decoder_for(encoding);
-        VERIFY(decoder);
+        VERIFY(decoder.has_value());
 
         StringBuilder builder { input.length() };
         bool last_was_carriage_return = false;

Userland/Libraries/LibWeb/Encoding/TextDecoder.cpp

@@ -15,7 +15,7 @@ namespace Web::Encoding {
 WebIDL::ExceptionOr<JS::NonnullGCPtr<TextDecoder>> TextDecoder::construct_impl(JS::Realm& realm, DeprecatedFlyString encoding)
 {
     auto decoder = TextCodec::decoder_for(encoding);
-    if (!decoder)
+    if (!decoder.has_value())
         return WebIDL::SimpleException { WebIDL::SimpleExceptionType::TypeError, DeprecatedString::formatted("Invalid encoding {}", encoding) };
 
     return MUST_OR_THROW_OOM(realm.heap().allocate<TextDecoder>(realm, realm, *decoder, move(encoding), false, false));

Userland/Libraries/LibWeb/HTML/HTMLScriptElement.cpp

@@ -514,7 +514,7 @@ void HTMLScriptElement::resource_did_load()
     // If the resource has an explicit encoding (i.e from a HTTP Content-Type header)
     // we have to re-encode it to UTF-8.
     if (resource()->has_encoding()) {
-        if (auto* codec = TextCodec::decoder_for(resource()->encoding().value())) {
+        if (auto codec = TextCodec::decoder_for(resource()->encoding().value()); codec.has_value()) {
             data = codec->to_utf8(data).to_byte_buffer();
         }
     }

Userland/Libraries/LibWeb/HTML/Parser/HTMLTokenizer.cpp

@@ -2798,8 +2798,8 @@ HTMLTokenizer::HTMLTokenizer()
 
 HTMLTokenizer::HTMLTokenizer(StringView input, DeprecatedString const& encoding)
 {
-    auto* decoder = TextCodec::decoder_for(encoding);
+    auto decoder = TextCodec::decoder_for(encoding);
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
     m_decoded_input = decoder->to_utf8(input);
     m_utf8_view = Utf8View(m_decoded_input);
     m_utf8_iterator = m_utf8_view.begin();

Userland/Libraries/LibWeb/HTML/Window.cpp

@@ -1420,7 +1420,7 @@ JS_DEFINE_NATIVE_FUNCTION(Window::atob)
     // The bytes object might contain bytes greater than 128, encode them in UTF8
     // NOTE: Any 8-bit encoding -> utf-8 decoder will work for this
     auto text_decoder = TextCodec::decoder_for("windows-1252"sv);
-    VERIFY(text_decoder);
+    VERIFY(text_decoder.has_value());
     auto text = text_decoder->to_utf8(decoded.release_value());
 
     return JS::PrimitiveString::create(vm, DeprecatedString(text));

Userland/Libraries/LibWeb/HTML/WorkerGlobalScope.cpp

@@ -161,8 +161,8 @@ WebIDL::ExceptionOr<DeprecatedString> WorkerGlobalScope::atob(DeprecatedString c
 
     // 3. Return decodedData.
     // decode_base64() returns a byte string. LibJS uses UTF-8 for strings. Use Latin1Decoder to convert bytes 128-255 to UTF-8.
-    auto* decoder = TextCodec::decoder_for("windows-1252"sv);
+    auto decoder = TextCodec::decoder_for("windows-1252"sv);
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
     return decoder->to_utf8(decoded_data.value());
 }
 

Userland/Libraries/LibWeb/Loader/Resource.cpp

@@ -83,7 +83,7 @@ static DeprecatedString mime_type_from_content_type(DeprecatedString const& cont
 
 static bool is_valid_encoding(StringView encoding)
 {
-    return TextCodec::decoder_for(encoding);
+    return TextCodec::decoder_for(encoding).has_value();
 }
 
 void Resource::did_load(Badge<ResourceLoader>, ReadonlyBytes data, HashMap<DeprecatedString, DeprecatedString, CaseInsensitiveStringTraits> const& headers, Optional<u32> status_code)

Userland/Libraries/LibWeb/XHR/XMLHttpRequest.cpp

@@ -219,10 +219,10 @@ DeprecatedString XMLHttpRequest::get_text_response() const
         charset = "UTF-8"sv;
 
     // 5. Return the result of running decode on xhr’s received bytes using fallback encoding charset.
-    auto* decoder = TextCodec::decoder_for(charset.value());
+    auto decoder = TextCodec::decoder_for(charset.value());
 
     // If we don't support the decoder yet, let's crash instead of attempting to return something, as the result would be incorrect and create obscure bugs.
-    VERIFY(decoder);
+    VERIFY(decoder.has_value());
 
     return TextCodec::convert_input_to_utf8_using_given_decoder_unless_there_is_a_byte_order_mark(*decoder, m_received_bytes);
 }

Userland/Utilities/js.cpp

@@ -882,8 +882,8 @@ ErrorOr<int> serenity_main(Main::Arguments arguments)
                 if (Utf8View { file_contents }.validate()) {
                     builder.append(source);
                 } else {
-                    auto* decoder = TextCodec::decoder_for("windows-1252"sv);
+                    auto decoder = TextCodec::decoder_for("windows-1252"sv);
-                    VERIFY(decoder);
+                    VERIFY(decoder.has_value());
 
                     auto utf8_source = TextCodec::convert_input_to_utf8_using_given_decoder_unless_there_is_a_byte_order_mark(*decoder, source);
                     builder.append(utf8_source);