LibXML+LibWeb: Use existing HTML entities table for XML parsing too

For XHTML documents, resolve named character entities (e.g.,  )
using the HTML entity table via a getEntity SAX callback. This avoids
parsing a large embedded DTD on every document and matches the approach
used by Blink and WebKit.

This also removes the now-unused DTD infrastructure:

- Remove resolve_external_resource callback from Parser::Options
- Remove resolve_xml_resource() function and its ~60KB embedded DTD
- Remove all call sites passing the unused callback

Libraries/LibWeb/DOM/DocumentLoading.cpp

@@ -56,7 +56,7 @@ bool build_xml_document(DOM::Document& document, ByteBuffer const& data, Optiona
         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 });
+    XML::Parser parser(source, { .resolve_named_html_entity = resolve_named_html_entity });
     XMLDocumentBuilder builder { document };
     auto result = parser.parse_with_listener(builder);
     return !result.is_error() && !builder.has_error();
@@ -189,7 +189,7 @@ static WebIDL::ExceptionOr<GC::Ref<DOM::Document>> load_xml_document(HTML::Navig
         }
         // NB: If document is part of session history traversal, resolve the signal_to_continue_session_history_processing.
         signal_to_continue_session_history_processing->resolve({});
-        XML::Parser parser(source.value(), { .preserve_cdata = true, .preserve_comments = true, .resolve_external_resource = resolve_xml_resource });
+        XML::Parser parser(source.value(), { .preserve_cdata = true, .preserve_comments = true, .resolve_named_html_entity = resolve_named_html_entity });
         XMLDocumentBuilder builder { document };
         auto result = parser.parse_with_listener(builder);
         if (result.is_error()) {

Libraries/LibWeb/HTML/DOMParser.cpp

@@ -72,7 +72,7 @@ WebIDL::ExceptionOr<GC::Root<DOM::Document>> DOMParser::parse_from_string(Truste
 
         // 1. Create an XML parser parse, associated with document, and with XML scripting support disabled.
         auto const utf8_complaint_string = compliant_string.to_utf8_but_should_be_ported_to_utf16();
-        XML::Parser parser(utf8_complaint_string, { .resolve_external_resource = resolve_xml_resource });
+        XML::Parser parser(utf8_complaint_string, { .resolve_named_html_entity = resolve_named_html_entity });
         XMLDocumentBuilder builder { *document, XMLScriptingSupport::Disabled };
         // 2. Parse compliantString using parser.
         auto result = parser.parse_with_listener(builder);

Libraries/LibWeb/SVG/SVGDecodedImageData.cpp

@@ -62,7 +62,7 @@ ErrorOr<GC::Ref<SVGDecodedImageData>> SVGDecodedImageData::create(JS::Realm& rea
     auto& window = as<HTML::Window>(HTML::relevant_global_object(document));
     document->browsing_context()->window_proxy()->set_window(window);
 
-    XML::Parser parser(data, { .resolve_external_resource = resolve_xml_resource });
+    XML::Parser parser(data, { .resolve_named_html_entity = resolve_named_html_entity });
     XMLDocumentBuilder builder { document };
     auto result = parser.parse_with_listener(builder);
     (void)result;

Libraries/LibWeb/XML/XMLDocumentBuilder.h

@@ -21,7 +21,7 @@ enum class XMLScriptingSupport {
     Enabled,
 };
 
-ErrorOr<Variant<ByteString, Vector<XML::MarkupDeclaration>>> resolve_xml_resource(XML::SystemID const&, Optional<XML::PublicID> const&);
+Optional<String> resolve_named_html_entity(StringView entity_name);
 
 class XMLDocumentBuilder final : public XML::Listener {
 public:

Libraries/LibWeb/XML/XMLFragmentParser.cpp

@@ -72,7 +72,7 @@ WebIDL::ExceptionOr<Vector<GC::Root<DOM::Node>>> XMLFragmentParser::parse_xml_fr
     if (allow_declarative_shadow_roots == HTML::HTMLParser::AllowDeclarativeShadowRoots::Yes)
         document->set_allow_declarative_shadow_roots(true);
 
-    XML::Parser parser(feed.string_view());
+    XML::Parser parser(feed.string_view(), { .resolve_named_html_entity = resolve_named_html_entity });
     XMLDocumentBuilder builder { *document, XMLScriptingSupport::Disabled };
     auto result = parser.parse_with_listener(builder);
 

Libraries/LibXML/Parser/Parser.cpp

@@ -29,6 +29,8 @@ struct ParserContext {
 
     Vector<ParseError> parse_errors;
 
+    Parser::Options const* options { nullptr };
+    bool is_xhtml_document { false };
     int depth { 0 };
 };
 
@@ -53,6 +55,82 @@ static StringView xml_char_to_string_view(xmlChar const* str)
     return StringView(reinterpret_cast<char const*>(str), strlen(reinterpret_cast<char const*>(str)));
 }
 
+static bool is_known_xhtml_public_id(StringView public_id)
+{
+    return public_id.is_one_of(
+        "-//W3C//DTD XHTML 1.0 Transitional//EN"sv,
+        "-//W3C//DTD XHTML 1.1//EN"sv,
+        "-//W3C//DTD XHTML 1.0 Strict//EN"sv,
+        "-//W3C//DTD XHTML 1.0 Frameset//EN"sv,
+        "-//W3C//DTD XHTML Basic 1.0//EN"sv,
+        "-//W3C//DTD XHTML 1.1 plus MathML 2.0//EN"sv,
+        "-//W3C//DTD XHTML 1.1 plus MathML 2.0 plus SVG 1.1//EN"sv,
+        "-//W3C//DTD MathML 2.0//EN"sv,
+        "-//WAPFORUM//DTD XHTML Mobile 1.0//EN"sv,
+        "-//WAPFORUM//DTD XHTML Mobile 1.1//EN"sv,
+        "-//WAPFORUM//DTD XHTML Mobile 1.2//EN"sv);
+}
+
+static void external_subset_handler(void* ctx, xmlChar const*, xmlChar const* external_id, xmlChar const*)
+{
+    auto* parser_ctx = static_cast<xmlParserCtxtPtr>(ctx);
+    auto* context = static_cast<ParserContext*>(parser_ctx->_private);
+    if (!context || !external_id)
+        return;
+
+    auto public_id = xml_char_to_string_view(external_id);
+    if (is_known_xhtml_public_id(public_id))
+        context->is_xhtml_document = true;
+}
+
+static xmlEntity s_xhtml_entity_result;
+static char s_xhtml_entity_utf8_buffer[32];
+
+static xmlEntityPtr get_entity_handler(void* ctx, xmlChar const* name)
+{
+    auto* parser_ctx = static_cast<xmlParserCtxtPtr>(ctx);
+
+    auto* predefined = xmlGetPredefinedEntity(name);
+    if (predefined)
+        return predefined;
+
+    if (parser_ctx->myDoc) {
+        auto* doc_entity = xmlGetDocEntity(parser_ctx->myDoc, name);
+        if (doc_entity)
+            return doc_entity;
+    }
+
+    auto* context = static_cast<ParserContext*>(parser_ctx->_private);
+    if (!context || !context->is_xhtml_document)
+        return nullptr;
+
+    // For XHTML documents, resolve named character entities (e.g., &nbsp;) using the
+    // HTML entity table. This avoids parsing a large embedded DTD on every document
+    // and matches the approach used by Blink and WebKit.
+    if (!context->options || !context->options->resolve_named_html_entity)
+        return nullptr;
+
+    auto entity_name = xml_char_to_string_view(name);
+    auto resolved = context->options->resolve_named_html_entity(entity_name);
+    if (!resolved.has_value())
+        return nullptr;
+
+    auto utf8_bytes = resolved->bytes_as_string_view();
+    if (utf8_bytes.length() >= sizeof(s_xhtml_entity_utf8_buffer))
+        return nullptr;
+
+    (void)utf8_bytes.copy_characters_to_buffer(s_xhtml_entity_utf8_buffer, sizeof(s_xhtml_entity_utf8_buffer));
+
+    s_xhtml_entity_result = {};
+    s_xhtml_entity_result.type = XML_ENTITY_DECL;
+    s_xhtml_entity_result.name = name;
+    s_xhtml_entity_result.content = reinterpret_cast<xmlChar*>(s_xhtml_entity_utf8_buffer);
+    s_xhtml_entity_result.length = static_cast<int>(utf8_bytes.length());
+    s_xhtml_entity_result.etype = XML_INTERNAL_PREDEFINED_ENTITY;
+
+    return &s_xhtml_entity_result;
+}
+
 static void start_document_handler(void* ctx)
 {
     auto* parser_ctx = static_cast<xmlParserCtxtPtr>(ctx);
@@ -345,7 +423,7 @@ static void structured_error_handler(void* ctx, xmlError const* error)
         context->error = move(parse_error);
 }
 
-static xmlSAXHandler create_sax_handler(bool preserve_comments)
+static xmlSAXHandler create_sax_handler(bool preserve_comments, bool resolve_html_entities)
 {
     xmlSAXHandler handler = {};
     handler.initialized = XML_SAX2_MAGIC;
@@ -360,6 +438,10 @@ static xmlSAXHandler create_sax_handler(bool preserve_comments)
     handler.serror = structured_error_handler;
     if (preserve_comments)
         handler.comment = comment_handler;
+    if (resolve_html_entities) {
+        handler.externalSubset = external_subset_handler;
+        handler.getEntity = get_entity_handler;
+    }
     return handler;
 }
 
@@ -371,8 +453,10 @@ ErrorOr<void, ParseError> Parser::parse_with_listener(Listener& listener)
 
     ParserContext context;
     context.listener = &listener;
+    context.options = &m_options;
 
-    auto sax_handler = create_sax_handler(m_options.preserve_comments);
+    bool resolve_html_entities = static_cast<bool>(m_options.resolve_named_html_entity);
+    auto sax_handler = create_sax_handler(m_options.preserve_comments, resolve_html_entities);
 
     int options = XML_PARSE_NONET | XML_PARSE_NOWARNING;
     if (!m_options.preserve_cdata)
@@ -412,8 +496,10 @@ ErrorOr<void, ParseError> Parser::parse_with_listener(Listener& listener)
 ErrorOr<Document, ParseError> Parser::parse()
 {
     ParserContext context;
+    context.options = &m_options;
 
-    auto sax_handler = create_sax_handler(m_options.preserve_comments);
+    bool resolve_html_entities = static_cast<bool>(m_options.resolve_named_html_entity);
+    auto sax_handler = create_sax_handler(m_options.preserve_comments, resolve_html_entities);
 
     int options = XML_PARSE_NONET | XML_PARSE_NOWARNING;
     if (!m_options.preserve_cdata)

Libraries/LibXML/Parser/Parser.h

@@ -11,7 +11,9 @@
 #include <AK/Function.h>
 #include <AK/GenericLexer.h>
 #include <AK/HashMap.h>
+#include <AK/Optional.h>
 #include <AK/OwnPtr.h>
+#include <AK/String.h>
 #include <LibXML/DOM/Document.h>
 #include <LibXML/DOM/DocumentTypeDeclaration.h>
 #include <LibXML/DOM/Node.h>
@@ -51,7 +53,7 @@ public:
         bool preserve_cdata { true };
         bool preserve_comments { false };
         bool treat_errors_as_fatal { true };
-        Function<ErrorOr<Variant<ByteString, Vector<MarkupDeclaration>>>(SystemID const&, Optional<PublicID> const&)> resolve_external_resource {};
+        Function<Optional<String>(StringView)> resolve_named_html_entity {};
     };
 
     Parser(StringView source, Options options)

Tests/LibWeb/Ref/expected/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/adopt-while-parsing-001-ref.html

@@ -0,0 +1,9 @@
+<!doctype html>
+<title>Test reference</title>
+<style>
+  html, body { margin: 0 }
+</style>
+<iframe src="about:blank"></iframe>
+<div>
+  PASS
+</div>

Tests/LibWeb/Ref/expected/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/data-xhtml-with-dtd-ref.html

@@ -0,0 +1,4 @@
+<!DOCTYPE html>
+Test passes if it correctly shows &Aacute; in the subframe.
+<hr>
+<iframe srcdoc="&amp;Aacute"></iframe>

Tests/LibWeb/Ref/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/adopt-while-parsing-001.html

@@ -0,0 +1,19 @@
+<!doctype html>
+<title>Appending from the parser after adopting in an XML document doesn't miss notifications</title>
+<link rel="match" href="../../../../../expected/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/adopt-while-parsing-001-ref.html">
+<link rel="help" href="https://bugzilla.mozilla.org/show_bug.cgi?id=1511329">
+<link rel="author" title="Emilio Cobos Álvarez" href="mailto:emilio@crisal.io">
+<link rel="author" title="Mozilla" href="https://mozilla.org">
+<style>
+  html, body { margin: 0 }
+</style>
+<script>
+  // If we don't get notified of the <div> insertion, the PASS text will never appear.
+  function parsingInterrupted() {
+    let frameDoc = document.querySelector("iframe").contentDocument;
+    let root = frameDoc.documentElement;
+    document.documentElement.appendChild(root);
+    root.offsetTop;
+  }
+</script>
+<iframe src="support/adopt-while-parsing.xhtml"></iframe>

Tests/LibWeb/Ref/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/data-xhtml-with-dtd.html

@@ -0,0 +1,27 @@
+<!DOCTYPE html>
+<!--
+    Any copyright is dedicated to the Public Domain.
+    http://creativecommons.org/publicdomain/zero/1.0/
+  -->
+<html>
+  <head>
+    <title>
+      Test that an XHTML document with a data: URL still handles the XHTML DTD
+      properly even if the DTD URL is given as a relative URL.
+    </title>
+    <link rel="author" title="Boris Zbarsky" href="bzbarsky@mit.edu">
+    <link rel="match" href="../../../../../expected/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/data-xhtml-with-dtd-ref.html">
+  </head>
+  <body>
+    Test passes if it correctly shows &Aacute; in the subframe.
+    <hr>
+    <!-- Document in the subframe is:
+<?xml version="1.0"?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml">
+  <body>
+    &Aacute;
+  </body>
+</html>
+    -->
+    <iframe src='data:application/xml,%3C%3Fxml%20version%3D%221.0%22%3F%3E%0A%3C!DOCTYPE%20html%20PUBLIC%20%22-%2F%2FW3C%2F%2FDTD%20XHTML%201.0%20Strict%2F%2FEN%22%20%22DTD%2Fxhtml1-strict.dtd%22%3E%0A%3Chtml%20xmlns%3D%22http%3A%2F%2Fwww.w3.org%2F1999%2Fxhtml%22%3E%0A%20%20%3Cbody%3E%0A%20%20%20%20%26Aacute%3B%0A%20%20%3C%2Fbody%3E%0A%3C%2Fhtml%3E%0A'></iframe>

Tests/LibWeb/Ref/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/support/adopt-while-parsing.xhtml

@@ -0,0 +1,11 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml">
+<body>
+  <script>
+    window.parent.parsingInterrupted();
+  </script>
+  <div>
+    PASS
+  </div>
+</body>
+</html>

Tests/LibWeb/Text/expected/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xml-stylesheet-blocking.txt

@@ -0,0 +1,6 @@
+Harness status: OK
+
+Found 1 tests
+
+1 Fail
+Fail	xml-stylesheet blocks script execution and rendering

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/support/simple-style.css

@@ -0,0 +1 @@
+:root { z-index: 3 }

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/support/xhtml-mathml-dtd-entity.htm

@@ -0,0 +1,49 @@
+<!DOCTYPE html>
+<meta charset=utf-8>
+<script>
+  var parser = new DOMParser();
+  var parse = parser.parseFromString.bind(parser);
+
+  function generateTestFunction(entitystring, expectedString, publicId, systemId, mimeType, friendlyMime) {
+    return function () {
+      var doctypeString = '<!DOCTYPE html';
+      if (publicId != null)
+        doctypeString += ' PUBLIC "' + publicId + '" "' + systemId + '">';
+      else if (systemId != null)
+        doctypeString += ' SYSTEM "' + systemId + '">';
+      else // both are null
+        doctypeString += '>';
+      var doc = parse(doctypeString + "<html><head></head><body id='test'>"+entitystring+"</body></html>", mimeType);
+      var root = doc.getElementById('test');
+      parent.assert_not_equals(root, null, friendlyMime + " parsing the entity reference caused a parse error;");
+      parent.assert_true(!!root.firstChild);
+      // Next line because some browsers include the partial parsed result in the parser error returned document.
+      parent.assert_equals(root.firstChild.nodeType, 3/*Text*/, friendlyMime + " parsing the entity reference caused a parse error;");
+      var text = root.firstChild.data;
+      for (var i = 0, len = expectedString.length; i < len; i++) {
+        parent.assert_equals(text.charCodeAt(i),expectedString.charCodeAt(i));
+      }
+    }
+  }
+
+  function setupTests(jsonEntities, publicId, systemId, mimeType, friendlyMime) {
+    for (entityName in jsonEntities) {
+      if ((mimeType == "text/html") || /;$/.test(entityName)) {
+        parent.test(generateTestFunction(entityName, jsonEntities[entityName].characters, publicId, systemId, mimeType, friendlyMime), friendlyMime + " parsing " + entityName);
+      }
+    }
+  }
+
+  parent.setup(function() {}, {explicit_done: true});
+
+  function run(row) {
+    var xhr = new XMLHttpRequest();
+    xhr.open("GET", "entities.json");
+    xhr.onload = function () {
+      var entitiesJSON = JSON.parse(xhr.response);
+      setupTests(entitiesJSON, row[1], row[2], row[0], row[3]);
+      parent.done();
+    }
+    xhr.send();
+  }
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-1.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD XHTML 1.0 Transitional//EN", "foo", "XHTML1.0 Transitional"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-10.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["text/html", null, null, "HTML"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-11.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//WAPFORUM//DTD XHTML Mobile 1.1//EN", "foo", "XHTML Mobile 1.1"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-12.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//WAPFORUM//DTD XHTML Mobile 1.2//EN", "foo", "XHTML Mobile 1.2"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-2.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD XHTML 1.1//EN", "foo", "XHTML1.1"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-3.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD XHTML 1.0 Strict//EN", "foo", "XHTML1.0 Strict"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-4.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD XHTML 1.0 Frameset//EN", "foo", "XHTML1.0 Frameset"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-5.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD XHTML Basic 1.0//EN", "foo", "XHTML Basic"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-6.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD XHTML 1.1 plus MathML 2.0//EN", "foo", "XHTML1.1+MathML"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-7.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD XHTML 1.1 plus MathML 2.0 plus SVG 1.1//EN", "foo", "XHTML1.1+MathML+SVG"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-8.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//W3C//DTD MathML 2.0//EN", "foo", "MathML"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-9.htm

@@ -0,0 +1,13 @@
+<!DOCTYPE html>
+<meta name=timeout content=long>
+<title>HTML entities for various XHTML Doctype</title>
+<link rel=help href="http://w3c.github.io/html/xhtml.html#parsing-xhtml-documents">
+
+<script src="../../../resources/testharness.js"></script>
+<script src="../../../resources/testharnessreport.js"></script>
+<iframe id="test" src="support/xhtml-mathml-dtd-entity.htm"></iframe>
+
+<script>
+onload = () => document.getElementById("test").contentWindow.run(
+["application/xhtml+xml", "-//WAPFORUM//DTD XHTML Mobile 1.0//EN", "foo", "XHTML Mobile"]);
+</script>

Tests/LibWeb/Text/input/wpt-import/html/the-xhtml-syntax/parsing-xhtml-documents/xml-stylesheet-blocking.xhtml

@@ -0,0 +1,22 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<?xml-stylesheet type="text/css" href="support/simple-style.css?pipe=trickle(d2)"?>
+<html xmlns="http://www.w3.org/1999/xhtml">
+  <head>
+    <script><![CDATA[
+      window.observedZIndex = getComputedStyle(document.documentElement).zIndex;
+    ]]></script>
+    <title>xml-stylesheet blocks script execution and rendering</title>
+    <link rel="help" href="https://bugzilla.mozilla.org/show_bug.cgi?id=1986042" />
+    <link rel="author" title="Emilio Cobos Álvarez" href="mailto:emilio@crisal.io" />
+    <link rel="author" title="Mozilla" href="https://mozilla.org" />
+    <script src="../../../resources/testharness.js"/>
+    <script src="../../../resources/testharnessreport.js"/>
+  </head>
+  <body>
+    <script><![CDATA[
+    test(function() {
+      assert_equals(window.observedZIndex, "3", "XML processing instruction should've blocked script execution and rendering");
+    });
+    ]]></script>
+  </body>
+</html>

Utilities/xml.cpp

@@ -362,18 +362,6 @@ static auto parse(StringView contents)
         contents,
         {
             .preserve_comments = true,
-            .resolve_external_resource = [&](XML::SystemID const& system_id, Optional<XML::PublicID> const&) -> ErrorOr<Variant<ByteString, Vector<XML::MarkupDeclaration>>> {
-                auto base = URL::create_with_file_scheme(s_path);
-                auto url = URL::Parser::basic_parse(system_id.system_literal, base);
-                if (!url.has_value())
-                    return Error::from_string_literal("Invalid URL");
-
-                if (url->scheme() != "file")
-                    return Error::from_string_literal("NYI: Nonlocal entity");
-
-                auto file = TRY(Core::File::open(URL::percent_decode(url->serialize_path()), Core::File::OpenMode::Read));
-                return ByteString::copy(TRY(file->read_until_eof()));
-            },
         },
     };
 }