LibTextCodec+LibWeb: Move isomorphic coders to LibTextCodec

trflynn89 · gmta · commit 0fd80a8f9926 · 2025-11-27T14:57:29.000+01:00
This will be used outside of LibWeb.
diff --git a/Libraries/LibTextCodec/Decoder.cpp b/Libraries/LibTextCodec/Decoder.cpp
@@ -1299,4 +1299,18 @@ ErrorOr<void> ReplacementDecoder::process(StringView input, Function<ErrorOr<voi
     return {};
 }
 
+// https://infra.spec.whatwg.org/#isomorphic-decode
+String isomorphic_decode(StringView input)
+{
+    // To isomorphic decode a byte sequence input, return a string whose code point length is equal to input’s length
+    // and whose code points have the same values as the values of input’s bytes, in the same order.
+    // NB: This is essentially spec-speak for "Decode as ISO-8859-1 / Latin-1".
+    StringBuilder builder(input.length());
+
+    for (auto byte : input.bytes())
+        builder.append_code_point(byte);
+
+    return builder.to_string_without_validation();
+}
+
 }
diff --git a/Libraries/LibTextCodec/Decoder.h b/Libraries/LibTextCodec/Decoder.h
@@ -135,4 +135,6 @@ TEXTCODEC_API ErrorOr<String> convert_input_to_utf8_using_given_decoder_unless_t
 
 TEXTCODEC_API StringView get_output_encoding(StringView encoding);
 
+TEXTCODEC_API String isomorphic_decode(StringView);
+
 }
diff --git a/Libraries/LibTextCodec/Encoder.cpp b/Libraries/LibTextCodec/Encoder.cpp
@@ -667,4 +667,23 @@ ErrorOr<void> SingleByteEncoder<ArrayType>::process(Utf8View input, Function<Err
     return {};
 }
 
+// https://infra.spec.whatwg.org/#isomorphic-encode
+ByteString isomorphic_encode(StringView input)
+{
+    // To isomorphic encode an isomorphic string input: return a byte sequence whose length is equal to input’s code
+    // point length and whose bytes have the same values as the values of input’s code points, in the same order.
+    // NB: This is essentially spec-speak for "Encode as ISO-8859-1 / Latin-1".
+    StringBuilder builder(input.length());
+
+    for (auto code_point : Utf8View { input }) {
+        // VERIFY(code_point <= 0xFF);
+        if (code_point > 0xFF)
+            dbgln("FIXME: Trying to isomorphic encode a string with code points > U+00FF.");
+
+        builder.append(static_cast<u8>(code_point));
+    }
+
+    return builder.to_byte_string();
+}
+
 }
diff --git a/Libraries/LibTextCodec/Encoder.h b/Libraries/LibTextCodec/Encoder.h
@@ -90,4 +90,6 @@ class SingleByteEncoder final : public Encoder {
 TEXTCODEC_API Optional<Encoder&> encoder_for_exact_name(StringView encoding);
 TEXTCODEC_API Optional<Encoder&> encoder_for(StringView label);
 
+TEXTCODEC_API ByteString isomorphic_encode(StringView);
+
 }
diff --git a/Libraries/LibWeb/ContentSecurityPolicy/Policy.cpp b/Libraries/LibWeb/ContentSecurityPolicy/Policy.cpp
@@ -6,6 +6,7 @@
 
 #include <AK/GenericLexer.h>
 #include <AK/String.h>
+#include <LibTextCodec/Decoder.h>
 #include <LibWeb/ContentSecurityPolicy/Directives/DirectiveFactory.h>
 #include <LibWeb/ContentSecurityPolicy/Directives/SerializedDirective.h>
 #include <LibWeb/ContentSecurityPolicy/Policy.h>
@@ -14,7 +15,6 @@
 #include <LibWeb/Fetch/Infrastructure/HTTP/Headers.h>
 #include <LibWeb/Fetch/Infrastructure/HTTP/Responses.h>
 #include <LibWeb/Infra/CharacterTypes.h>
-#include <LibWeb/Infra/Strings.h>
 
 namespace Web::ContentSecurityPolicy {
 
@@ -31,7 +31,7 @@ GC::Ref<Policy> Policy::parse_a_serialized_csp(GC::Heap& heap, Variant<ByteStrin
     // 1. If serialized is a byte sequence, then set serialized to be the result of isomorphic decoding serialized.
     auto serialized_string = serialized.has<String>()
         ? serialized.get<String>()
-        : Infra::isomorphic_decode(serialized.get<ByteString>());
+        : TextCodec::isomorphic_decode(serialized.get<ByteString>());
 
     // 2. Let policy be a new policy with an empty directive set, a source of source, and a disposition of disposition.
     auto policy = heap.allocate<Policy>();
diff --git a/Libraries/LibWeb/DOM/Document.cpp b/Libraries/LibWeb/DOM/Document.cpp
@@ -23,6 +23,7 @@
 #include <LibJS/Runtime/Array.h>
 #include <LibJS/Runtime/FunctionObject.h>
 #include <LibJS/Runtime/NativeFunction.h>
+#include <LibTextCodec/Decoder.h>
 #include <LibURL/Origin.h>
 #include <LibURL/Parser.h>
 #include <LibUnicode/Segmenter.h>
@@ -425,7 +426,7 @@ WebIDL::ExceptionOr<GC::Ref<Document>> Document::create_and_initialize(Type type
     // 15. If navigationParams's response has a `Refresh` header, then:
     if (auto maybe_refresh = navigation_params.response->header_list()->get("Refresh"sv); maybe_refresh.has_value()) {
         // 1. Let value be the isomorphic decoding of the value of the header.
-        auto value = Infra::isomorphic_decode(maybe_refresh.value());
+        auto value = TextCodec::isomorphic_decode(maybe_refresh.value());
 
         // 2. Run the shared declarative refresh steps with document and value.
         document->shared_declarative_refresh_steps(value, nullptr);
diff --git a/Libraries/LibWeb/Fetch/Body.cpp b/Libraries/LibWeb/Fetch/Body.cpp
@@ -24,7 +24,6 @@
 #include <LibWeb/FileAPI/File.h>
 #include <LibWeb/HTML/Scripting/TemporaryExecutionContext.h>
 #include <LibWeb/Infra/JSON.h>
-#include <LibWeb/Infra/Strings.h>
 #include <LibWeb/MimeSniff/MimeType.h>
 #include <LibWeb/Streams/ReadableStream.h>
 #include <LibWeb/WebIDL/Promise.h>
@@ -353,7 +352,7 @@ static MultipartParsingErrorOr<MultiPartFormDataHeader> parse_multipart_form_dat
             header_value = header_value.trim(Infrastructure::HTTP_TAB_OR_SPACE, TrimMode::Right);
 
             // 3. Set contentType to the isomorphic decoding of header value.
-            header.content_type = Infra::isomorphic_decode(header_value.bytes());
+            header.content_type = TextCodec::isomorphic_decode(header_value);
         }
         // -> Otherwise
         else {
diff --git a/Libraries/LibWeb/Fetch/Fetching/Fetching.cpp b/Libraries/LibWeb/Fetch/Fetching/Fetching.cpp
@@ -14,6 +14,7 @@
 #include <AK/ScopeGuard.h>
 #include <LibJS/Runtime/Completion.h>
 #include <LibRequests/RequestTimingInfo.h>
+#include <LibTextCodec/Encoder.h>
 #include <LibWeb/Bindings/MainThreadVM.h>
 #include <LibWeb/Bindings/PrincipalHostDefined.h>
 #include <LibWeb/ContentSecurityPolicy/BlockingAlgorithms.h>
@@ -51,7 +52,6 @@
 #include <LibWeb/HTML/Window.h>
 #include <LibWeb/HTML/WorkerGlobalScope.h>
 #include <LibWeb/HighResolutionTime/TimeOrigin.h>
-#include <LibWeb/Infra/Strings.h>
 #include <LibWeb/Loader/LoadRequest.h>
 #include <LibWeb/Loader/ResourceLoader.h>
 #include <LibWeb/MixedContent/AbstractOperations.h>
@@ -1831,7 +1831,7 @@ GC::Ref<PendingResponse> http_network_or_cache_fetch(JS::Realm& realm, Infrastru
         // 11. If httpRequest’s referrer is a URL, then:
         if (auto const* referrer_url = http_request->referrer().get_pointer<URL::URL>()) {
             // 1. Let referrerValue be httpRequest’s referrer, serialized and isomorphic encoded.
-            auto referrer_value = Infra::isomorphic_encode(referrer_url->serialize());
+            auto referrer_value = TextCodec::isomorphic_encode(referrer_url->serialize());
 
             // 2. Append (`Referer`, referrerValue) to httpRequest’s header list.
             http_request->header_list()->append({ "Referer"sv, move(referrer_value) });
diff --git a/Libraries/LibWeb/Fetch/Headers.cpp b/Libraries/LibWeb/Fetch/Headers.cpp
@@ -6,10 +6,10 @@
 
 #include <LibJS/Runtime/Completion.h>
 #include <LibJS/Runtime/VM.h>
+#include <LibTextCodec/Decoder.h>
 #include <LibWeb/Bindings/HeadersPrototype.h>
 #include <LibWeb/Bindings/Intrinsics.h>
 #include <LibWeb/Fetch/Headers.h>
-#include <LibWeb/Infra/Strings.h>
 
 namespace Web::Fetch {
 
@@ -102,7 +102,7 @@ WebIDL::ExceptionOr<Optional<String>> Headers::get(String const& name)
 
     // 2. Return the result of getting name from this’s header list.
     auto byte_buffer = m_header_list->get(name);
-    return byte_buffer.has_value() ? Infra::isomorphic_decode(*byte_buffer) : Optional<String> {};
+    return byte_buffer.has_value() ? TextCodec::isomorphic_decode(*byte_buffer) : Optional<String> {};
 }
 
 // https://fetch.spec.whatwg.org/#dom-headers-getsetcookie
@@ -119,7 +119,7 @@ Vector<String> Headers::get_set_cookie()
     //    `Set-Cookie`, in order.
     for (auto const& header : *m_header_list) {
         if (header.name.equals_ignoring_ascii_case("Set-Cookie"sv))
-            values.append(Infra::isomorphic_decode(header.value));
+            values.append(TextCodec::isomorphic_decode(header.value));
     }
     return values;
 }
@@ -187,7 +187,7 @@ JS::ThrowCompletionOr<void> Headers::for_each(ForEachCallback callback)
         auto const& pair = pairs[i];
 
         // 2. Invoke idlCallback with « pair’s value, pair’s key, idlObject » and with thisArg as the callback this value.
-        TRY(callback(Infra::isomorphic_decode(pair.name), Infra::isomorphic_decode(pair.value)));
+        TRY(callback(TextCodec::isomorphic_decode(pair.name), TextCodec::isomorphic_decode(pair.value)));
 
         // 3. Set pairs to idlObject’s current list of value pairs to iterate over. (It might have changed.)
         pairs = value_pairs_to_iterate_over();
diff --git a/Libraries/LibWeb/Fetch/HeadersIterator.cpp b/Libraries/LibWeb/Fetch/HeadersIterator.cpp
@@ -6,10 +6,10 @@
 
 #include <LibJS/Runtime/Array.h>
 #include <LibJS/Runtime/Iterator.h>
+#include <LibTextCodec/Decoder.h>
 #include <LibWeb/Bindings/HeadersIteratorPrototype.h>
 #include <LibWeb/Bindings/Intrinsics.h>
 #include <LibWeb/Fetch/HeadersIterator.h>
-#include <LibWeb/Infra/Strings.h>
 
 namespace Web::Bindings {
 
@@ -66,8 +66,8 @@ GC::Ref<JS::Object> HeadersIterator::next()
         return create_iterator_result_object(vm(), JS::js_undefined(), true);
 
     auto const& pair = pairs[m_index++];
-    auto pair_name = Infra::isomorphic_decode(pair.name);
-    auto pair_value = Infra::isomorphic_decode(pair.value);
+    auto pair_name = TextCodec::isomorphic_decode(pair.name);
+    auto pair_value = TextCodec::isomorphic_decode(pair.value);
 
     switch (m_iteration_kind) {
     case JS::Object::PropertyKind::Key:
diff --git a/Libraries/LibWeb/Fetch/Infrastructure/HTTP/Headers.cpp b/Libraries/LibWeb/Fetch/Infrastructure/HTTP/Headers.cpp
@@ -14,10 +14,10 @@
 #include <LibJS/Runtime/VM.h>
 #include <LibRegex/Regex.h>
 #include <LibTextCodec/Decoder.h>
+#include <LibTextCodec/Encoder.h>
 #include <LibWeb/Fetch/Infrastructure/HTTP.h>
 #include <LibWeb/Fetch/Infrastructure/HTTP/Headers.h>
 #include <LibWeb/Fetch/Infrastructure/HTTP/Methods.h>
-#include <LibWeb/Infra/Strings.h>
 #include <LibWeb/Loader/ResourceLoader.h>
 #include <LibWeb/MimeSniff/MimeType.h>
 
@@ -28,8 +28,8 @@ GC_DEFINE_ALLOCATOR(HeaderList);
 Header Header::isomorphic_encode(StringView name, StringView value)
 {
     return {
-        .name = Infra::isomorphic_encode(name),
-        .value = Infra::isomorphic_encode(value),
+        .name = TextCodec::isomorphic_encode(name),
+        .value = TextCodec::isomorphic_encode(value),
     };
 }
 
@@ -523,7 +523,7 @@ Vector<String> get_decode_and_split_header_value(StringView value)
     // To get, decode, and split a header value value, run these steps:
 
     // 1. Let input be the result of isomorphic decoding value.
-    auto input = Infra::isomorphic_decode(value);
+    auto input = TextCodec::isomorphic_decode(value);
 
     // 2. Let position be a position variable for input, initially pointing at the start of input.
     GenericLexer lexer { input };
@@ -612,7 +612,7 @@ ByteString build_content_range(u64 range_start, u64 range_end, u64 full_length)
 Optional<RangeHeaderValue> parse_single_range_header_value(StringView const value, bool const allow_whitespace)
 {
     // 1. Let data be the isomorphic decoding of value.
-    auto const data = Infra::isomorphic_decode(value);
+    auto const data = TextCodec::isomorphic_decode(value);
 
     // 2. If data does not start with "bytes", then return failure.
     if (!data.starts_with_bytes("bytes"sv))
@@ -709,7 +709,7 @@ bool is_cors_safelisted_request_header(Header const& header)
             return false;
 
         // 2. Let mimeType be the result of parsing the result of isomorphic decoding value.
-        auto decoded = Infra::isomorphic_decode(value);
+        auto decoded = TextCodec::isomorphic_decode(value);
         auto mime_type = MimeSniff::MimeType::parse(decoded);
 
         // 3. If mimeType is failure, then return false.
diff --git a/Libraries/LibWeb/Fetch/Infrastructure/HTTP/Requests.cpp b/Libraries/LibWeb/Fetch/Infrastructure/HTTP/Requests.cpp
@@ -7,14 +7,14 @@
 #include <AK/Array.h>
 #include <LibGC/Heap.h>
 #include <LibJS/Runtime/Realm.h>
+#include <LibTextCodec/Encoder.h>
 #include <LibWeb/ContentSecurityPolicy/Directives/Names.h>
 #include <LibWeb/ContentSecurityPolicy/PolicyList.h>
 #include <LibWeb/ContentSecurityPolicy/Violation.h>
 #include <LibWeb/DOMURL/DOMURL.h>
 #include <LibWeb/Fetch/Fetching/PendingResponse.h>
 #include <LibWeb/Fetch/Infrastructure/HTTP/Requests.h>
 #include <LibWeb/HTML/TraversableNavigable.h>
-#include <LibWeb/Infra/Strings.h>
 
 namespace Web::Fetch::Infrastructure {
 
@@ -217,7 +217,7 @@ ByteString Request::byte_serialize_origin() const
 {
     // Byte-serializing a request origin, given a request request, is to return the result of serializing a request
     // origin with request, isomorphic encoded.
-    return Infra::isomorphic_encode(serialize_origin());
+    return TextCodec::isomorphic_encode(serialize_origin());
 }
 
 // https://fetch.spec.whatwg.org/#concept-request-clone
diff --git a/Libraries/LibWeb/Fetch/Infrastructure/URL.cpp b/Libraries/LibWeb/Fetch/Infrastructure/URL.cpp
@@ -7,10 +7,10 @@
  */
 
 #include <AK/Base64.h>
+#include <LibTextCodec/Decoder.h>
 #include <LibURL/URL.h>
 #include <LibWeb/Export.h>
 #include <LibWeb/Fetch/Infrastructure/URL.h>
-#include <LibWeb/Infra/Strings.h>
 #include <LibWeb/MimeSniff/MimeType.h>
 
 namespace Web::Fetch::Infrastructure {
@@ -77,7 +77,7 @@ ErrorOr<DataURL> process_data_url(URL::URL const& data_url)
         trimmed_substring_view = trimmed_substring_view.trim(" "sv, TrimMode::Right);
         if (trimmed_substring_view.ends_with(';')) {
             // 1. Let stringBody be the isomorphic decode of body.
-            auto string_body = Infra::isomorphic_decode(body);
+            auto string_body = TextCodec::isomorphic_decode(body);
 
             // 2. Set body to the forgiving-base64 decode of stringBody.
             // 3. If body is failure, then return failure.
diff --git a/Libraries/LibWeb/HTML/UniversalGlobalScope.cpp b/Libraries/LibWeb/HTML/UniversalGlobalScope.cpp
@@ -13,13 +13,13 @@
 #include <AK/Vector.h>
 #include <LibGC/Function.h>
 #include <LibJS/Runtime/NativeFunction.h>
+#include <LibTextCodec/Decoder.h>
 #include <LibWeb/HTML/PromiseRejectionEvent.h>
 #include <LibWeb/HTML/Scripting/ExceptionReporter.h>
 #include <LibWeb/HTML/StructuredSerialize.h>
 #include <LibWeb/HTML/StructuredSerializeOptions.h>
 #include <LibWeb/HTML/UniversalGlobalScope.h>
 #include <LibWeb/HTML/Window.h>
-#include <LibWeb/Infra/Strings.h>
 #include <LibWeb/WebIDL/AbstractOperations.h>
 #include <LibWeb/WebIDL/DOMException.h>
 #include <LibWeb/WebIDL/ExceptionOr.h>
@@ -71,7 +71,7 @@ WebIDL::ExceptionOr<String> UniversalGlobalScopeMixin::atob(String const& data)
 
     // 3. Return decodedData.
     // decode_base64() returns a byte buffer. LibJS uses UTF-8 for strings. Use isomorphic decoding to convert bytes to UTF-8.
-    return Infra::isomorphic_decode(decoded_data.value());
+    return TextCodec::isomorphic_decode(decoded_data.value());
 }
 
 // https://html.spec.whatwg.org/multipage/timers-and-user-prompts.html#dom-queuemicrotask
diff --git a/Libraries/LibWeb/Infra/Strings.cpp b/Libraries/LibWeb/Infra/Strings.cpp
@@ -156,39 +156,6 @@ ErrorOr<String> convert_to_scalar_value_string(StringView string)
     return scalar_value_builder.to_string();
 }
 
-// https://infra.spec.whatwg.org/#isomorphic-encode
-ByteString isomorphic_encode(StringView input)
-{
-    // To isomorphic encode an isomorphic string input: return a byte sequence whose length is equal to input’s code
-    // point length and whose bytes have the same values as the values of input’s code points, in the same order.
-    // NB: This is essentially spec-speak for "Encode as ISO-8859-1 / Latin-1".
-    StringBuilder builder(input.length());
-
-    for (auto code_point : Utf8View { input }) {
-        // VERIFY(code_point <= 0xFF);
-        if (code_point > 0xFF)
-            dbgln("FIXME: Trying to isomorphic encode a string with code points > U+00FF.");
-
-        builder.append(static_cast<u8>(code_point));
-    }
-
-    return builder.to_byte_string();
-}
-
-// https://infra.spec.whatwg.org/#isomorphic-decode
-String isomorphic_decode(StringView input)
-{
-    // To isomorphic decode a byte sequence input, return a string whose code point length is equal to input’s length
-    // and whose code points have the same values as the values of input’s bytes, in the same order.
-    // NB: This is essentially spec-speak for "Decode as ISO-8859-1 / Latin-1".
-    StringBuilder builder(input.length());
-
-    for (auto byte : input.bytes())
-        builder.append_code_point(byte);
-
-    return builder.to_string_without_validation();
-}
-
 // https://infra.spec.whatwg.org/#code-unit-less-than
 bool code_unit_less_than(StringView a, StringView b)
 {
diff --git a/Libraries/LibWeb/Infra/Strings.h b/Libraries/LibWeb/Infra/Strings.h
@@ -20,8 +20,6 @@ WEB_API ErrorOr<String> strip_and_collapse_whitespace(StringView string);
 Utf16String strip_and_collapse_whitespace(Utf16String const& string);
 WEB_API bool is_code_unit_prefix(StringView potential_prefix, StringView input);
 WEB_API ErrorOr<String> convert_to_scalar_value_string(StringView string);
-ByteString isomorphic_encode(StringView input);
-WEB_API String isomorphic_decode(StringView input);
 bool code_unit_less_than(StringView a, StringView b);
 
 }
diff --git a/Libraries/LibWeb/XHR/XMLHttpRequest.cpp b/Libraries/LibWeb/XHR/XMLHttpRequest.cpp
@@ -43,7 +43,6 @@
 #include <LibWeb/HTML/Window.h>
 #include <LibWeb/Infra/ByteSequences.h>
 #include <LibWeb/Infra/JSON.h>
-#include <LibWeb/Infra/Strings.h>
 #include <LibWeb/Loader/ResourceLoader.h>
 #include <LibWeb/Page/Page.h>
 #include <LibWeb/Platform/EventLoopPlugin.h>
@@ -968,7 +967,7 @@ Optional<String> XMLHttpRequest::get_response_header(String const& name) const
         return {};
 
     // FIXME: The spec doesn't mention isomorphic decode. Spec bug?
-    return Infra::isomorphic_decode(header_bytes->bytes());
+    return TextCodec::isomorphic_decode(*header_bytes);
 }
 
 // https://xhr.spec.whatwg.org/#legacy-uppercased-byte-less-than
@@ -1002,7 +1001,7 @@ String XMLHttpRequest::get_all_response_headers() const
     // 4. For each header in headers, append header’s name, followed by a 0x3A 0x20 byte pair, followed by header’s value, followed by a 0x0D 0x0A byte pair, to output.
     for (auto const& header : initial_headers) {
         // FIXME: The spec does not mention isomorphic decode. Spec bug?
-        output.appendff("{}: {}\r\n", header.name, Infra::isomorphic_decode(header.value));
+        output.appendff("{}: {}\r\n", header.name, TextCodec::isomorphic_decode(header.value));
     }
 
     // 5. Return output.

Original file line number	Diff line number	Diff line change
`@@ -135,4 +135,6 @@ TEXTCODEC_API ErrorOr<String> convert_input_to_utf8_using_given_decoder_unless_t`
`135`	`135`
`136`	`136`	`TEXTCODEC_API StringView get_output_encoding(StringView encoding);`
`137`	`137`
	`138`	`+TEXTCODEC_API String isomorphic_decode(StringView);`
	`139`	`+`
`138`	`140`	`}`
Original file line number	Diff line number	Diff line change
`@@ -90,4 +90,6 @@ class SingleByteEncoder final : public Encoder {`
`90`	`90`	`TEXTCODEC_API Optional<Encoder&> encoder_for_exact_name(StringView encoding);`
`91`	`91`	`TEXTCODEC_API Optional<Encoder&> encoder_for(StringView label);`
`92`	`92`
	`93`	`+TEXTCODEC_API ByteString isomorphic_encode(StringView);`
	`94`	`+`
`93`	`95`	`}`