1
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2025-06-08 05:27:14 +09:00
ladybird/Libraries/LibWeb/Encoding/TextDecoder.cpp
Timothy Flynn 0a256b0a9a AK+Everywhere: Change StringView case conversions to return String
There's a bit of a UTF-8 assumption with this change. But nearly every
caller of these methods were immediately creating a String from the
resulting ByteString anyways.
2025-04-07 17:44:38 +02:00

85 lines
3.4 KiB
C++
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

/*
* Copyright (c) 2022, Ali Mohammad Pur <mpfard@serenityos.org>
* Copyright (c) 2024, Simon Wanner <simon@skyrising.xyz>
*
* SPDX-License-Identifier: BSD-2-Clause
*/
#include <AK/FlyString.h>
#include <LibJS/Runtime/TypedArray.h>
#include <LibWeb/Bindings/Intrinsics.h>
#include <LibWeb/Bindings/TextDecoderPrototype.h>
#include <LibWeb/Encoding/TextDecoder.h>
#include <LibWeb/WebIDL/AbstractOperations.h>
#include <LibWeb/WebIDL/Buffers.h>
namespace Web::Encoding {
GC_DEFINE_ALLOCATOR(TextDecoder);
// https://encoding.spec.whatwg.org/#dom-textdecoder
WebIDL::ExceptionOr<GC::Ref<TextDecoder>> TextDecoder::construct_impl(JS::Realm& realm, FlyString label, Optional<TextDecoderOptions> const& options)
{
auto& vm = realm.vm();
// 1. Let encoding be the result of getting an encoding from label.
auto encoding = TextCodec::get_standardized_encoding(label);
// 2. If encoding is failure or replacement, then throw a RangeError.
if (!encoding.has_value() || encoding->equals_ignoring_ascii_case("replacement"sv))
return WebIDL::SimpleException { WebIDL::SimpleExceptionType::RangeError, TRY_OR_THROW_OOM(vm, String::formatted("Invalid encoding {}", label)) };
// 3. Set thiss encoding to encoding.
// https://encoding.spec.whatwg.org/#dom-textdecoder-encoding
// The encoding getter steps are to return thiss encodings name, ASCII lowercased.
auto lowercase_encoding_name = encoding.value().to_ascii_lowercase_string();
// 4. If options["fatal"] is true, then set thiss error mode to "fatal".
auto fatal = options.value_or({}).fatal;
// 5. Set thiss ignore BOM to options["ignoreBOM"].
auto ignore_bom = options.value_or({}).ignore_bom;
// NOTE: This should happen in decode(), but we don't support streaming yet and share decoders across calls.
auto decoder = TextCodec::decoder_for_exact_name(encoding.value());
VERIFY(decoder.has_value());
return realm.create<TextDecoder>(realm, *decoder, lowercase_encoding_name, fatal, ignore_bom);
}
// https://encoding.spec.whatwg.org/#dom-textdecoder
TextDecoder::TextDecoder(JS::Realm& realm, TextCodec::Decoder& decoder, FlyString encoding, bool fatal, bool ignore_bom)
: PlatformObject(realm)
, m_decoder(decoder)
, m_encoding(move(encoding))
, m_fatal(fatal)
, m_ignore_bom(ignore_bom)
{
}
TextDecoder::~TextDecoder() = default;
void TextDecoder::initialize(JS::Realm& realm)
{
Base::initialize(realm);
WEB_SET_PROTOTYPE_FOR_INTERFACE(TextDecoder);
}
// https://encoding.spec.whatwg.org/#dom-textdecoder-decode
WebIDL::ExceptionOr<String> TextDecoder::decode(Optional<GC::Root<WebIDL::BufferSource>> const& input, Optional<TextDecodeOptions> const&) const
{
if (!input.has_value())
return TRY_OR_THROW_OOM(vm(), m_decoder.to_utf8({}));
// FIXME: Implement the streaming stuff.
auto data_buffer_or_error = WebIDL::get_buffer_source_copy(*input.value()->raw_object());
if (data_buffer_or_error.is_error())
return WebIDL::OperationError::create(realm(), "Failed to copy bytes from ArrayBuffer"_string);
auto& data_buffer = data_buffer_or_error.value();
auto result = TRY_OR_THROW_OOM(vm(), m_decoder.to_utf8({ data_buffer.data(), data_buffer.size() }));
if (this->fatal() && result.contains(0xfffd))
return WebIDL::SimpleException { WebIDL::SimpleExceptionType::TypeError, "Decoding failed"sv };
return result;
}
}