LibWeb: Store HTTP methods and headers as ByteString

The spec declares these as a byte sequence, which we then implemented as
a ByteBuffer. This has become pretty awkward to deal with, as evidenced
by the plethora of `MUST(ByteBuffer::copy(...))` and `.bytes()` calls
everywhere inside Fetch. We would then treat the bytes as a string
anyways by wrapping them in StringView everywhere.

We now store these as a ByteString. This is more comfortable to deal
with, and we no longer need to continually copy underlying storage (as
ByteString is ref-counted).

This work is largely preparatory for an upcoming HTTP header refactor.

Author: trflynn89

Libraries/LibWeb/ContentSecurityPolicy/Policy.cpp

@@ -21,7 +21,7 @@ namespace Web::ContentSecurityPolicy {
 GC_DEFINE_ALLOCATOR(Policy);
 
 // https://w3c.github.io/webappsec-csp/#abstract-opdef-parse-a-serialized-csp
-GC::Ref<Policy> Policy::parse_a_serialized_csp(GC::Heap& heap, Variant<ByteBuffer, String> serialized, Source source, Disposition disposition)
+GC::Ref<Policy> Policy::parse_a_serialized_csp(GC::Heap& heap, Variant<ByteString, String> serialized, Source source, Disposition disposition)
 {
     // To parse a serialized CSP, given a byte sequence or string serialized, a source source, and a disposition disposition,
     // execute the following steps.
@@ -31,7 +31,7 @@ GC::Ref<Policy> Policy::parse_a_serialized_csp(GC::Heap& heap, Variant<ByteBuffe
     // 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<ByteBuffer>());
+        : Infra::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>();
@@ -101,9 +101,9 @@ GC::Ref<PolicyList> Policy::parse_a_responses_content_security_policies(GC::Heap
 
     // 2. For each token returned by extracting header list values given Content-Security-Policy and response’s header
     //    list:
-    auto enforce_policy_tokens_or_failure = response->header_list()->extract_header_list_values("Content-Security-Policy"sv.bytes());
+    auto enforce_policy_tokens_or_failure = response->header_list()->extract_header_list_values("Content-Security-Policy"sv);
 
-    if (auto const* enforce_policy_tokens = enforce_policy_tokens_or_failure.get_pointer<Vector<ByteBuffer>>()) {
+    if (auto const* enforce_policy_tokens = enforce_policy_tokens_or_failure.get_pointer<Vector<ByteString>>()) {
         for (auto const& enforce_policy_token : *enforce_policy_tokens) {
             // 1. Let policy be the result of parsing token, with a source of "header", and a disposition of "enforce".
             auto policy = parse_a_serialized_csp(heap, enforce_policy_token, Policy::Source::Header, Policy::Disposition::Enforce);
@@ -116,9 +116,9 @@ GC::Ref<PolicyList> Policy::parse_a_responses_content_security_policies(GC::Heap
 
     // 3. For each token returned by extracting header list values given Content-Security-Policy-Report-Only and
     //    response’s header list:
-    auto report_policy_tokens_or_failure = response->header_list()->extract_header_list_values("Content-Security-Policy-Report-Only"sv.bytes());
+    auto report_policy_tokens_or_failure = response->header_list()->extract_header_list_values("Content-Security-Policy-Report-Only"sv);
 
-    if (auto const* report_policy_tokens = enforce_policy_tokens_or_failure.get_pointer<Vector<ByteBuffer>>()) {
+    if (auto const* report_policy_tokens = enforce_policy_tokens_or_failure.get_pointer<Vector<ByteString>>()) {
         for (auto const& report_policy_token : *report_policy_tokens) {
             // 1. Let policy be the result of parsing token, with a source of "header", and a disposition of "report".
             auto policy = parse_a_serialized_csp(heap, report_policy_token, Policy::Source::Header, Policy::Disposition::Report);

Libraries/LibWeb/ContentSecurityPolicy/Policy.h

@@ -38,7 +38,7 @@ class Policy final : public GC::Cell {
 
     ~Policy() = default;
 
-    [[nodiscard]] static GC::Ref<Policy> parse_a_serialized_csp(GC::Heap&, Variant<ByteBuffer, String> serialized, Source source, Disposition disposition);
+    [[nodiscard]] static GC::Ref<Policy> parse_a_serialized_csp(GC::Heap&, Variant<ByteString, String> serialized, Source source, Disposition disposition);
     [[nodiscard]] static GC::Ref<PolicyList> parse_a_responses_content_security_policies(GC::Heap&, GC::Ref<Fetch::Infrastructure::Response const> response);
     [[nodiscard]] static GC::Ref<Policy> create_from_serialized_policy(GC::Heap&, SerializedPolicy const&);
 

Libraries/LibWeb/ContentSecurityPolicy/Violation.cpp

@@ -29,7 +29,7 @@ GC_DEFINE_ALLOCATOR(Violation);
 Violation::Violation(GC::Ptr<JS::Object> global_object, GC::Ref<Policy const> policy, String directive)
     : m_global_object(global_object)
     , m_policy(policy)
-    , m_effective_directive(directive)
+    , m_effective_directive(move(directive))
 {
 }
 
@@ -38,7 +38,7 @@ GC::Ref<Violation> Violation::create_a_violation_object_for_global_policy_and_di
 {
     // 1. Let violation be a new violation whose global object is global, policy is policy, effective directive is
     //    directive, and resource is null.
-    auto violation = realm.create<Violation>(global_object, policy, directive);
+    auto violation = realm.create<Violation>(global_object, policy, move(directive));
 
     // FIXME: 2. If the user agent is currently executing script, and can extract a source file’s URL, line number,
     //           and column number from the global, set violation’s source file, line number, and column number
@@ -406,7 +406,7 @@ void Violation::report_a_violation(JS::Realm& realm)
 
                         // method
                         //    "POST"
-                        request->set_method(MUST(ByteBuffer::copy("POST"sv.bytes())));
+                        request->set_method("POST"sv);
 
                         // url
                         //    violation’s url
@@ -448,8 +448,7 @@ void Violation::report_a_violation(JS::Realm& realm)
                         //    A header list containing a single header whose name is "Content-Type", and value is
                         //    "application/csp-report"
                         auto header_list = Fetch::Infrastructure::HeaderList::create(vm);
-                        auto content_type_header = Fetch::Infrastructure::Header::from_string_pair("Content-Type"sv, "application/csp-report"sv);
-                        header_list->append(move(content_type_header));
+                        header_list->append({ "Content-Type"sv, "application/csp-report"sv });
                         request->set_header_list(header_list);
 
                         // body

Libraries/LibWeb/DOM/Document.cpp

@@ -385,15 +385,15 @@ WebIDL::ExceptionOr<GC::Ref<Document>> Document::create_and_initialize(Type type
     document->m_window = window;
 
     // NOTE: Non-standard: Pull out the Last-Modified header for use in the lastModified property.
-    if (auto maybe_last_modified = navigation_params.response->header_list()->get("Last-Modified"sv.bytes()); maybe_last_modified.has_value()) {
+    if (auto maybe_last_modified = navigation_params.response->header_list()->get("Last-Modified"sv); maybe_last_modified.has_value()) {
         // rfc9110, 8.8.2: The Last-Modified header field must be in GMT time zone.
         // document->m_last_modified is in local time zone.
         document->m_last_modified = AK::UnixDateTime::parse("%a, %d %b %Y %H:%M:%S GMT"sv, maybe_last_modified.value(), true);
     }
 
     // NOTE: Non-standard: Pull out the Content-Language header to determine the document's language.
-    if (auto maybe_http_content_language = navigation_params.response->header_list()->get("Content-Language"sv.bytes()); maybe_http_content_language.has_value()) {
-        if (auto maybe_content_language = String::from_utf8(maybe_http_content_language.value()); !maybe_content_language.is_error())
+    if (auto maybe_http_content_language = navigation_params.response->header_list()->get("Content-Language"sv); maybe_http_content_language.has_value()) {
+        if (auto maybe_content_language = String::from_byte_string(maybe_http_content_language.value()); !maybe_content_language.is_error())
             document->m_http_content_language = maybe_content_language.release_value();
     }
 
@@ -423,7 +423,7 @@ WebIDL::ExceptionOr<GC::Ref<Document>> Document::create_and_initialize(Type type
     //            navigationParams's response's service worker timing info.
 
     // 15. If navigationParams's response has a `Refresh` header, then:
-    if (auto maybe_refresh = navigation_params.response->header_list()->get("Refresh"sv.bytes()); maybe_refresh.has_value()) {
+    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());
 

Libraries/LibWeb/Fetch/BodyInit.cpp

@@ -72,7 +72,7 @@ WebIDL::ExceptionOr<Infrastructure::BodyWithType> extract_body(JS::Realm& realm,
     Optional<u64> length {};
 
     // 9. Let type be null.
-    Optional<ByteBuffer> type {};
+    Optional<ByteString> type {};
 
     // 10. Switch on object.
     TRY(object.visit(
@@ -83,7 +83,7 @@ WebIDL::ExceptionOr<Infrastructure::BodyWithType> extract_body(JS::Realm& realm,
             length = blob->size();
             // If object’s type attribute is not the empty byte sequence, set type to its value.
             if (!blob->type().is_empty())
-                type = MUST(ByteBuffer::copy(blob->type().bytes()));
+                type = blob->type().to_byte_string();
             return {};
         },
         [&](ReadonlyBytes bytes) -> WebIDL::ExceptionOr<void> {
@@ -103,24 +103,23 @@ WebIDL::ExceptionOr<Infrastructure::BodyWithType> extract_body(JS::Realm& realm,
             source = serialized_form_data.serialized_data;
             // FIXME: Set length to unclear, see html/6424 for improving this.
             // Set type to `multipart/form-data; boundary=`, followed by the multipart/form-data boundary string generated by the multipart/form-data encoding algorithm.
-            auto type_string = MUST(String::formatted("multipart/form-data; boundary={}", serialized_form_data.boundary));
-            type = MUST(ByteBuffer::copy(type_string.bytes()));
+            type = ByteString::formatted("multipart/form-data; boundary={}", serialized_form_data.boundary);
             return {};
         },
         [&](GC::Root<DOMURL::URLSearchParams> const& url_search_params) -> WebIDL::ExceptionOr<void> {
             // Set source to the result of running the application/x-www-form-urlencoded serializer with object’s list.
             auto search_params_string = url_search_params->to_string();
             source = MUST(ByteBuffer::copy(search_params_string.bytes()));
             // Set type to `application/x-www-form-urlencoded;charset=UTF-8`.
-            type = MUST(ByteBuffer::copy("application/x-www-form-urlencoded;charset=UTF-8"sv.bytes()));
+            type = "application/x-www-form-urlencoded;charset=UTF-8"sv;
             return {};
         },
         [&](String const& scalar_value_string) -> WebIDL::ExceptionOr<void> {
             // NOTE: AK::String is always UTF-8.
             // Set source to the UTF-8 encoding of object.
             source = MUST(ByteBuffer::copy(scalar_value_string.bytes()));
             // Set type to `text/plain;charset=UTF-8`.
-            type = MUST(ByteBuffer::copy("text/plain;charset=UTF-8"sv.bytes()));
+            type = "text/plain;charset=UTF-8"sv;
             return {};
         },
         [&](GC::Root<Streams::ReadableStream> const& stream) -> WebIDL::ExceptionOr<void> {
@@ -169,7 +168,7 @@ WebIDL::ExceptionOr<Infrastructure::BodyWithType> extract_body(JS::Realm& realm,
     auto body = Infrastructure::Body::create(vm, *stream, move(source), move(length));
 
     // 14. Return (body, type).
-    return Infrastructure::BodyWithType { .body = move(body), .type = move(type) };
+    return Infrastructure::BodyWithType { .body = body, .type = move(type) };
 }
 
 }

Libraries/LibWeb/Fetch/Fetching/Checks.cpp

@@ -15,15 +15,15 @@ namespace Web::Fetch::Fetching {
 bool cors_check(Infrastructure::Request const& request, Infrastructure::Response const& response)
 {
     // 1. Let origin be the result of getting `Access-Control-Allow-Origin` from response’s header list.
-    auto origin = response.header_list()->get("Access-Control-Allow-Origin"sv.bytes());
+    auto origin = response.header_list()->get("Access-Control-Allow-Origin"sv);
 
     // 2. If origin is null, then return failure.
     // NOTE: Null is not `null`.
     if (!origin.has_value())
         return false;
 
     // 3. If request’s credentials mode is not "include" and origin is `*`, then return success.
-    if (request.credentials_mode() != Infrastructure::Request::CredentialsMode::Include && origin->span() == "*"sv.bytes())
+    if (request.credentials_mode() != Infrastructure::Request::CredentialsMode::Include && *origin == "*"sv)
         return true;
 
     // 4. If the result of byte-serializing a request origin with request is not origin, then return failure.
@@ -35,10 +35,10 @@ bool cors_check(Infrastructure::Request const& request, Infrastructure::Response
         return true;
 
     // 6. Let credentials be the result of getting `Access-Control-Allow-Credentials` from response’s header list.
-    auto credentials = response.header_list()->get("Access-Control-Allow-Credentials"sv.bytes());
+    auto credentials = response.header_list()->get("Access-Control-Allow-Credentials"sv);
 
     // 7. If credentials is `true`, then return success.
-    if (credentials.has_value() && credentials->span() == "true"sv.bytes())
+    if (credentials == "true"sv)
         return true;
 
     // 8. Return failure.
@@ -53,7 +53,7 @@ bool tao_check(Infrastructure::Request const& request, Infrastructure::Response
         return false;
 
     // 2. Let values be the result of getting, decoding, and splitting `Timing-Allow-Origin` from response’s header list.
-    auto values = response.header_list()->get_decode_and_split("Timing-Allow-Origin"sv.bytes());
+    auto values = response.header_list()->get_decode_and_split("Timing-Allow-Origin"sv);
 
     // 3. If values contains "*", then return success.
     if (values.has_value() && values->contains_slow("*"sv))