LibWeb: Keep the tokens in ListOfActiveFormattingElements

Author: https://github.com/lpas
Commit: 6afd39b16a
Pull-request: https://github.com/LadybirdBrowser/ladybird/pull/6515
Reviewed-by: https://github.com/gmta ✅

Libraries/LibWeb/HTML/Parser/HTMLParser.cpp

@@ -2,6 +2,7 @@
  * Copyright (c) 2020-2025, Andreas Kling <andreas@ladybird.org>
  * Copyright (c) 2021, Luke Wilde <lukew@serenityos.org>
  * Copyright (c) 2023-2024, Shannon Booth <shannon@serenityos.org>
+ * Copyright (c) 2025, Lorenz Ackermann <me@lorenzackermann.xyz>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
@@ -1713,14 +1714,10 @@ Create:
     // 8. Create: Insert an HTML element for the token for which the element entry was created, to obtain new element.
     VERIFY(!entry->is_marker());
 
-    // FIXME: Hold on to the real token!
+    auto new_element = insert_html_element(*entry->token);
-    auto new_element = insert_html_element(HTMLToken::make_start_tag(entry->element->local_name()));
-    entry->element->for_each_attribute([&](auto& name, auto& value) {
-        new_element->append_attribute(name, value);
-    });
 
     // 9. Replace the entry for entry in the list with an entry for new element.
-    m_list_of_active_formatting_elements.entries().at(index).element = new_element;
+    m_list_of_active_formatting_elements.replace(*entry->element, new_element, *entry->token);
 
     // 10. If the entry for new element in the list of active formatting elements is not the last entry in the list, return to the step labeled advance.
     if (index != m_list_of_active_formatting_elements.entries().size() - 1)
@@ -1840,16 +1837,17 @@ HTMLParser::AdoptionAgencyAlgorithmOutcome HTMLParser::run_the_adoption_agency_a
                 break;
 
             // 4. If innerLoopCounter is greater than 3 and node is in the list of active formatting elements,
-            if (inner_loop_counter > 3 && m_list_of_active_formatting_elements.contains(*node)) {
+            auto node_index = m_list_of_active_formatting_elements.find_index(*node);
-                auto node_index = m_list_of_active_formatting_elements.find_index(*node);
+            if (inner_loop_counter > 3 && node_index.has_value()) {
-                if (node_index.has_value() && node_index.value() < bookmark)
+                if (node_index.value() < bookmark)
                     bookmark--;
                 // then remove node from the list of active formatting elements.
                 m_list_of_active_formatting_elements.remove(*node);
+                node_index = {};
             }
 
             // 5. If node is not in the list of active formatting elements,
-            if (!m_list_of_active_formatting_elements.contains(*node)) {
+            if (!node_index.has_value()) {
                 // then remove node from the stack of open elements and continue.
                 m_stack_of_open_elements.remove(*node);
                 continue;
@@ -1857,10 +1855,10 @@ HTMLParser::AdoptionAgencyAlgorithmOutcome HTMLParser::run_the_adoption_agency_a
 
             // 6. Create an element for the token for which the element node was created,
             //    in the HTML namespace, with commonAncestor as the intended parent;
-            // FIXME: hold onto the real token
+            auto& entry = m_list_of_active_formatting_elements.entries().at(node_index.value());
-            auto element = create_element_for(HTMLToken::make_start_tag(node->local_name()), Namespace::HTML, *common_ancestor);
+            auto element = create_element_for(*entry.token, Namespace::HTML, *common_ancestor);
             //    replace the entry for node in the list of active formatting elements with an entry for the new element,
-            m_list_of_active_formatting_elements.replace(*node, *element);
+            m_list_of_active_formatting_elements.replace(*node, *element, *entry.token);
             //    replace the entry for node in the stack of open elements with an entry for the new element,
             m_stack_of_open_elements.replace(*node, element);
             //    and let node be the new element.
@@ -1886,8 +1884,10 @@ HTMLParser::AdoptionAgencyAlgorithmOutcome HTMLParser::run_the_adoption_agency_a
 
         // 15. Create an element for the token for which formattingElement was created,
         //     in the HTML namespace, with furthestBlock as the intended parent.
-        // FIXME: hold onto the real token
+        auto formatting_element_index = m_list_of_active_formatting_elements.find_index(*formatting_element);
-        auto element = create_element_for(HTMLToken::make_start_tag(formatting_element->local_name()), Namespace::HTML, *furthest_block);
+        auto& entry = m_list_of_active_formatting_elements.entries().at(formatting_element_index.value());
+        auto token_data = ListOfActiveFormattingElements::create_own_token(*entry.token);
+        auto element = create_element_for(*token_data, Namespace::HTML, *furthest_block);
 
         // 16. Take all of the child nodes of furthestBlock and append them to the element created in the last step.
         for (auto& child : furthest_block->children_as_vector())
@@ -1898,11 +1898,10 @@ HTMLParser::AdoptionAgencyAlgorithmOutcome HTMLParser::run_the_adoption_agency_a
 
         // 18. Remove formattingElement from the list of active formatting elements,
         //     and insert the new element into the list of active formatting elements at the position of the aforementioned bookmark.
-        auto formatting_element_index = m_list_of_active_formatting_elements.find_index(*formatting_element);
         if (formatting_element_index.has_value() && formatting_element_index.value() < bookmark)
             bookmark--;
         m_list_of_active_formatting_elements.remove(*formatting_element);
-        m_list_of_active_formatting_elements.insert_at(bookmark, *element);
+        m_list_of_active_formatting_elements.insert_at(bookmark, *element, *token_data);
 
         // 19. Remove formattingElement from the stack of open elements, and insert the new element
         //     into the stack of open elements immediately below the position of furthestBlock in that stack.
@@ -2626,7 +2625,7 @@ void HTMLParser::handle_in_body(HTMLToken& token)
 
         // Insert an HTML element for the token. Push onto the list of active formatting elements that element.
         auto element = insert_html_element(token);
-        m_list_of_active_formatting_elements.add(*element);
+        m_list_of_active_formatting_elements.add(*element, token);
         return;
     }
 
@@ -2637,7 +2636,7 @@ void HTMLParser::handle_in_body(HTMLToken& token)
 
         // Insert an HTML element for the token. Push onto the list of active formatting elements that element.
         auto element = insert_html_element(token);
-        m_list_of_active_formatting_elements.add(*element);
+        m_list_of_active_formatting_elements.add(*element, token);
         return;
     }
 
@@ -2655,7 +2654,7 @@ void HTMLParser::handle_in_body(HTMLToken& token)
 
         // Insert an HTML element for the token. Push onto the list of active formatting elements that element.
         auto element = insert_html_element(token);
-        m_list_of_active_formatting_elements.add(*element);
+        m_list_of_active_formatting_elements.add(*element, token);
         return;
     }
 

Libraries/LibWeb/HTML/Parser/ListOfActiveFormattingElements.cpp

@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2020-2022, Andreas Kling <andreas@ladybird.org>
+ * Copyright (c) 2025, Lorenz Ackermann <me@lorenzackermann.xyz>
  *
  * SPDX-License-Identifier: BSD-2-Clause
  */
@@ -20,7 +21,7 @@ void ListOfActiveFormattingElements::visit_edges(JS::Cell::Visitor& visitor)
 
 void ListOfActiveFormattingElements::ensure_noahs_ark_clause(DOM::Element& element)
 {
-    Vector<Entry> possible_matches;
+    Vector<Entry&> possible_matches;
     for (size_t i = m_entries.size(); i > 0;) {
         i--;
         auto& entry = m_entries[i];
@@ -28,7 +29,7 @@ void ListOfActiveFormattingElements::ensure_noahs_ark_clause(DOM::Element& eleme
             break;
         if (entry.element->local_name() == element.local_name()
             && entry.element->namespace_uri() == element.namespace_uri()
-            && entry.element->attribute_list_size() == element.attribute_list_size())
+            && entry.token->attribute_count() == element.attribute_list_size())
             possible_matches.append(entry);
     }
 
@@ -38,7 +39,7 @@ void ListOfActiveFormattingElements::ensure_noahs_ark_clause(DOM::Element& eleme
     // FIXME: the attributes should be compared as they where created by the parser
     element.for_each_attribute([&](auto& name, auto& value) {
         possible_matches.remove_all_matching([&](auto& entry) {
-            auto attr = entry.element->get_attribute(name);
+            auto attr = entry.token->attribute(name);
             return !attr.has_value() || attr != value;
         });
     });
@@ -49,8 +50,19 @@ void ListOfActiveFormattingElements::ensure_noahs_ark_clause(DOM::Element& eleme
     remove(*possible_matches.last().element);
 }
 
+AK::OwnPtr<HTMLToken> ListOfActiveFormattingElements::create_own_token(HTMLToken& token)
+{
+    auto new_token = make<HTMLToken>(token.type());
+    new_token->set_tag_name(token.tag_name());
+    token.for_each_attribute([&](auto const& attribute) {
+        new_token->add_attribute(attribute);
+        return IterationDecision::Continue;
+    });
+    return new_token;
+}
+
 // https://html.spec.whatwg.org/multipage/parsing.html#push-onto-the-list-of-active-formatting-elements
-void ListOfActiveFormattingElements::add(DOM::Element& element)
+void ListOfActiveFormattingElements::add(DOM::Element& element, HTMLToken& token)
 {
     // 1. If there are already three elements in the list of active formatting elements after the last marker, if any, or anywhere in the list if there are no markers,
     //    that have the same tag name, namespace, and attributes as element, then remove the earliest such element from the list of active formatting elements.
@@ -58,12 +70,12 @@ void ListOfActiveFormattingElements::add(DOM::Element& element)
     //    can be paired such that the two attributes in each pair have identical names, namespaces, and values (the order of the attributes does not matter).
     ensure_noahs_ark_clause(element);
     // 2. Add element to the list of active formatting elements.
-    m_entries.append({ element });
+    m_entries.append({ element, create_own_token(token) });
 }
 
 void ListOfActiveFormattingElements::add_marker()
 {
-    m_entries.append({ nullptr });
+    m_entries.append({ nullptr, nullptr });
 }
 
 bool ListOfActiveFormattingElements::contains(DOM::Element const& element) const
@@ -112,17 +124,20 @@ Optional<size_t> ListOfActiveFormattingElements::find_index(DOM::Element const&
     return {};
 }
 
-void ListOfActiveFormattingElements::replace(DOM::Element& to_remove, DOM::Element& to_add)
+void ListOfActiveFormattingElements::replace(DOM::Element& to_remove, DOM::Element& to_add, HTMLToken& token)
 {
-    for (size_t i = 0; i < m_entries.size(); i++) {
+    for (auto& entry : m_entries) {
-        if (m_entries[i].element.ptr() == &to_remove)
+        if (entry.element.ptr() == &to_remove) {
-            m_entries[i].element = GC::make_root(to_add);
+            entry.element = GC::make_root(to_add);
+            entry.token = create_own_token(token);
+            break;
+        }
     }
 }
 
-void ListOfActiveFormattingElements::insert_at(size_t index, DOM::Element& element)
+void ListOfActiveFormattingElements::insert_at(size_t index, DOM::Element& element, HTMLToken& token)
 {
-    m_entries.insert(index, { element });
+    m_entries.insert(index, { element, create_own_token(token) });
 }
 
 }

Libraries/LibWeb/HTML/Parser/ListOfActiveFormattingElements.h

@@ -8,6 +8,7 @@
 
 #include <LibWeb/DOM/Element.h>
 #include <LibWeb/Forward.h>
+#include <LibWeb/HTML/Parser/HTMLToken.h>
 
 namespace Web::HTML {
 
@@ -20,16 +21,17 @@ public:
         bool is_marker() const { return !element; }
 
         GC::Ptr<DOM::Element> element;
+        AK::OwnPtr<HTMLToken> token;
     };
 
     bool is_empty() const { return m_entries.is_empty(); }
     bool contains(DOM::Element const&) const;
 
-    void add(DOM::Element& element);
+    void add(DOM::Element& element, HTMLToken& token);
     void add_marker();
-    void insert_at(size_t index, DOM::Element& element);
+    void insert_at(size_t index, DOM::Element& element, HTMLToken& token);
 
-    void replace(DOM::Element& to_remove, DOM::Element& to_add);
+    void replace(DOM::Element& to_remove, DOM::Element& to_add, HTMLToken& token);
 
     void remove(DOM::Element&);
 
@@ -44,6 +46,8 @@ public:
 
     void visit_edges(JS::Cell::Visitor&);
 
+    static AK::OwnPtr<HTMLToken> create_own_token(HTMLToken& token);
+
 private:
     Vector<Entry> m_entries;
     void ensure_noahs_ark_clause(DOM::Element& element);

Tests/LibWeb/Text/expected/wpt-import/html/syntax/parsing/html5lib_scripted_adoption01.txt

@@ -2,5 +2,5 @@ Harness status: OK
 
 Found 1 tests
 
-1 Fail
+1 Pass
-Fail	html5lib_scripted_adoption01.html 8970fe21b551a270aa74648bb2e8b905edb54522
+Pass	html5lib_scripted_adoption01.html 8970fe21b551a270aa74648bb2e8b905edb54522

Tests/LibWeb/Text/expected/wpt-import/html/syntax/parsing/html5lib_scripted_ark.txt

@@ -2,5 +2,5 @@ Harness status: OK
 
 Found 1 tests
 
-1 Fail
+1 Pass
-Fail	html5lib_scripted_ark.html b9a7cd0310cab4fd4eb77aed9149b966918e7ca2
+Pass	html5lib_scripted_ark.html b9a7cd0310cab4fd4eb77aed9149b966918e7ca2

Tests/LibWeb/Text/expected/wpt-import/html/syntax/parsing/html5lib_tricky01.txt

@@ -2,10 +2,9 @@ Harness status: OK
 
 Found 9 tests
 
-8 Pass
+9 Pass
-1 Fail
 Pass	html5lib_tricky01.html 06f0a6904729cd6a3ab91f3121c0b0eb54ee04d2
-Fail	html5lib_tricky01.html c99581b7d1d8c1cd421054891981c3fe8267e83c
+Pass	html5lib_tricky01.html c99581b7d1d8c1cd421054891981c3fe8267e83c
 Pass	html5lib_tricky01.html 09ba1d973acb46344442ea1e77a37de8736ce6e7
 Pass	html5lib_tricky01.html 9e40dd21a29521d60a43cb016f4100501ea26ec8
 Pass	html5lib_tricky01.html 23bed40fe77c77e3119528d9f77e041eeb77eebb