/* * Copyright (c) 2020, Andreas Kling * Copyright (c) 2023, Aliaksandr Kalenik * * SPDX-License-Identifier: BSD-2-Clause */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include namespace Web { static bool build_markdown_document(DOM::Document& document, ByteBuffer const& data) { auto markdown_document = Markdown::Document::parse(data); if (!markdown_document) return false; auto extra_head_contents = R"~~~( )~~~"sv; auto parser = HTML::HTMLParser::create(document, markdown_document->render_to_html(extra_head_contents), "utf-8"); parser->run(document.url()); return true; } static bool build_text_document(DOM::Document& document, ByteBuffer const& data) { auto html_element = DOM::create_element(document, HTML::TagNames::html, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(document.append_child(html_element)); auto head_element = DOM::create_element(document, HTML::TagNames::head, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(head_element)); auto title_element = DOM::create_element(document, HTML::TagNames::title, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(head_element->append_child(title_element)); auto title_text = document.create_text_node(MUST(String::from_deprecated_string(document.url().basename()))); MUST(title_element->append_child(title_text)); auto body_element = DOM::create_element(document, HTML::TagNames::body, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(body_element)); auto pre_element = DOM::create_element(document, HTML::TagNames::pre, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(body_element->append_child(pre_element)); MUST(pre_element->append_child(document.create_text_node(String::from_utf8(StringView { data }).release_value_but_fixme_should_propagate_errors()))); return true; } static bool build_image_document(DOM::Document& document, ByteBuffer const& data) { auto image = Platform::ImageCodecPlugin::the().decode_image(data); if (!image.has_value() || image->frames.is_empty()) return false; auto const& frame = image->frames[0]; auto const& bitmap = frame.bitmap; if (!bitmap) return false; auto html_element = DOM::create_element(document, HTML::TagNames::html, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(document.append_child(html_element)); auto head_element = DOM::create_element(document, HTML::TagNames::head, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(head_element)); auto title_element = DOM::create_element(document, HTML::TagNames::title, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(head_element->append_child(title_element)); auto basename = LexicalPath::basename(document.url().serialize_path()); auto title_text = document.heap().allocate(document.realm(), document, MUST(String::formatted("{} [{}x{}]", basename, bitmap->width(), bitmap->height()))); MUST(title_element->append_child(*title_text)); auto body_element = DOM::create_element(document, HTML::TagNames::body, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(body_element)); auto image_element = DOM::create_element(document, HTML::TagNames::img, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(image_element->set_attribute(HTML::AttributeNames::src, MUST(document.url().to_string()))); MUST(body_element->append_child(image_element)); return true; } static bool build_gemini_document(DOM::Document& document, ByteBuffer const& data) { StringView gemini_data { data }; auto gemini_document = Gemini::Document::parse(gemini_data, document.url()); DeprecatedString html_data = gemini_document->render_to_html(); dbgln_if(GEMINI_DEBUG, "Gemini data:\n\"\"\"{}\"\"\"", gemini_data); dbgln_if(GEMINI_DEBUG, "Converted to HTML:\n\"\"\"{}\"\"\"", html_data); auto parser = HTML::HTMLParser::create(document, html_data, "utf-8"); parser->run(document.url()); return true; } bool build_xml_document(DOM::Document& document, ByteBuffer const& data, Optional content_encoding) { Optional decoder; // The actual HTTP headers and other metadata, not the headers as mutated or implied by the algorithms given in this specification, // are the ones that must be used when determining the character encoding according to the rules given in the above specifications. if (content_encoding.has_value()) decoder = TextCodec::decoder_for(*content_encoding); if (!decoder.has_value()) { auto encoding = HTML::run_encoding_sniffing_algorithm(document, data); decoder = TextCodec::decoder_for(encoding); } VERIFY(decoder.has_value()); // Well-formed XML documents contain only properly encoded characters if (!decoder->validate(data)) return false; auto source = decoder->to_utf8(data).release_value_but_fixme_should_propagate_errors(); XML::Parser parser(source, { .resolve_external_resource = resolve_xml_resource }); XMLDocumentBuilder builder { document }; auto result = parser.parse_with_listener(builder); return !result.is_error() && !builder.has_error(); } static bool build_video_document(DOM::Document& document) { auto html_element = DOM::create_element(document, HTML::TagNames::html, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(document.append_child(html_element)); auto head_element = DOM::create_element(document, HTML::TagNames::head, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(head_element)); auto body_element = DOM::create_element(document, HTML::TagNames::body, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(body_element)); auto video_element = DOM::create_element(document, HTML::TagNames::video, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(video_element->set_attribute(HTML::AttributeNames::src, MUST(document.url().to_string()))); MUST(video_element->set_attribute(HTML::AttributeNames::autoplay, String {})); MUST(video_element->set_attribute(HTML::AttributeNames::controls, String {})); MUST(body_element->append_child(video_element)); return true; } static bool build_audio_document(DOM::Document& document) { auto html_element = DOM::create_element(document, HTML::TagNames::html, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(document.append_child(html_element)); auto head_element = DOM::create_element(document, HTML::TagNames::head, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(head_element)); auto body_element = DOM::create_element(document, HTML::TagNames::body, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(html_element->append_child(body_element)); auto video_element = DOM::create_element(document, HTML::TagNames::audio, Namespace::HTML).release_value_but_fixme_should_propagate_errors(); MUST(video_element->set_attribute(HTML::AttributeNames::src, MUST(document.url().to_string()))); MUST(video_element->set_attribute(HTML::AttributeNames::autoplay, String {})); MUST(video_element->set_attribute(HTML::AttributeNames::controls, String {})); MUST(body_element->append_child(video_element)); return true; } bool parse_document(DOM::Document& document, ByteBuffer const& data, Optional content_encoding) { auto& mime_type = document.content_type(); if (mime_type == "text/html") { auto parser = HTML::HTMLParser::create_with_uncertain_encoding(document, data); parser->run(document.url()); return true; } if (mime_type.ends_with_bytes("+xml"sv) || mime_type.is_one_of("text/xml", "application/xml")) return build_xml_document(document, data, move(content_encoding)); if (mime_type.starts_with_bytes("image/"sv)) return build_image_document(document, data); if (mime_type.starts_with_bytes("video/"sv)) return build_video_document(document); if (mime_type.starts_with_bytes("audio/"sv)) return build_audio_document(document); if (mime_type == "text/plain" || mime_type == "application/json") return build_text_document(document, data); if (mime_type == "text/markdown") return build_markdown_document(document, data); if (mime_type == "text/gemini") return build_gemini_document(document, data); return false; } static bool is_supported_document_mime_type(StringView mime_type) { if (mime_type == "text/html") return true; if (mime_type.ends_with("+xml"sv) || mime_type.is_one_of("text/xml", "application/xml")) return true; if (mime_type.starts_with("image/"sv)) return true; if (mime_type.starts_with("video/"sv)) return true; if (mime_type.starts_with("audio/"sv)) return true; if (mime_type == "text/plain" || mime_type == "application/json") return true; if (mime_type == "text/markdown") return true; if (mime_type == "text/gemini") return true; return false; } // https://html.spec.whatwg.org/multipage/browsing-the-web.html#loading-a-document JS::GCPtr load_document(Optional navigation_params) { VERIFY(navigation_params.has_value()); auto extracted_mime_type = navigation_params->response->header_list()->extract_mime_type().release_value_but_fixme_should_propagate_errors(); if (!extracted_mime_type.has_value()) return nullptr; auto mime_type = extracted_mime_type.release_value(); if (!is_supported_document_mime_type(mime_type.essence())) return nullptr; auto document = DOM::Document::create_and_initialize(DOM::Document::Type::HTML, "text/html", *navigation_params).release_value_but_fixme_should_propagate_errors(); document->set_content_type(mime_type.essence()); auto& realm = document->realm(); if (navigation_params->response->body()) { Optional content_encoding = mime_type.parameters().get("charset"sv); auto process_body = [document, url = navigation_params->response->url().value(), encoding = move(content_encoding)](ByteBuffer bytes) { if (parse_document(*document, bytes, move(encoding))) return; document->remove_all_children(true); auto error_html = load_error_page(url).release_value_but_fixme_should_propagate_errors(); auto parser = HTML::HTMLParser::create(document, error_html, "utf-8"); document->set_url(AK::URL("about:error")); parser->run(); }; auto process_body_error = [](auto) { dbgln("FIXME: Load html page with an error if read of body failed."); }; navigation_params->response->body()->fully_read( realm, move(process_body), move(process_body_error), JS::NonnullGCPtr { realm.global_object() }) .release_value_but_fixme_should_propagate_errors(); } return document; } // https://html.spec.whatwg.org/multipage/document-lifecycle.html#read-ua-inline JS::GCPtr create_document_for_inline_content(JS::GCPtr navigable, Optional navigation_id, StringView content_html) { auto& vm = navigable->vm(); // 1. Let origin be a new opaque origin. HTML::Origin origin {}; // 2. Let coop be a new cross-origin opener policy. auto coop = HTML::CrossOriginOpenerPolicy {}; // 3. Let coopEnforcementResult be a new cross-origin opener policy enforcement result with // url: response's URL // origin: origin // cross-origin opener policy: coop HTML::CrossOriginOpenerPolicyEnforcementResult coop_enforcement_result { .url = AK::URL("about:error"), // AD-HOC .origin = origin, .cross_origin_opener_policy = coop }; // 4. Let navigationParams be a new navigation params with // id: navigationId // navigable: navigable // request: null // response: a new response // origin: origin // fetch controller: null // commit early hints: null // COOP enforcement result: coopEnforcementResult // reserved environment: null // policy container: a new policy container // final sandboxing flag set: an empty set // cross-origin opener policy: coop // FIXME: navigation timing type: navTimingType // about base URL: null auto response = Fetch::Infrastructure::Response::create(vm); response->url_list().append(AK::URL("about:error")); // AD-HOC: https://github.com/whatwg/html/issues/9122 HTML::NavigationParams navigation_params { .id = navigation_id, .navigable = navigable, .request = {}, .response = *response, .fetch_controller = nullptr, .commit_early_hints = nullptr, .coop_enforcement_result = move(coop_enforcement_result), .reserved_environment = {}, .origin = move(origin), .policy_container = HTML::PolicyContainer {}, .final_sandboxing_flag_set = HTML::SandboxingFlagSet {}, .cross_origin_opener_policy = move(coop), .about_base_url = {}, }; // 5. Let document be the result of creating and initializing a Document object given "html", "text/html", and navigationParams. auto document = DOM::Document::create_and_initialize(DOM::Document::Type::HTML, "text/html", navigation_params).release_value_but_fixme_should_propagate_errors(); // 6. Either associate document with a custom rendering that is not rendered using the normal Document rendering rules, or mutate document until it represents the content the // user agent wants to render. auto parser = HTML::HTMLParser::create(document, content_html, "utf-8"); document->set_url(AK::URL("about:error")); parser->run(); // 7. Return document. return document; } }