LibWeb: Add basic support for :lang() CSS selector

This doesn't have parsing support for multiple languages in the same
selector. Support for language subcodes is not great either. But it
does do the basics.

Userland/Libraries/LibWeb/CSS/Parser/Parser.cpp

@@ -603,6 +603,14 @@ Result<Selector::SimpleSelector, Parser::ParsingResult> Parser::parse_simple_sel
                     return ParsingResult::SyntaxError;
                 }
                 simple_selector.pseudo_class.argument_selector_list = not_selector.release_value();
+            } else if (pseudo_function.name().equals_ignoring_case("lang"sv)) {
+                simple_selector.pseudo_class.type = Selector::SimpleSelector::PseudoClass::Type::Lang;
+                if (pseudo_function.values().is_empty()) {
+                    dbgln_if(CSS_PARSER_DEBUG, "Empty :lang() selector");
+                    return ParsingResult::SyntaxError;
+                }
+                // FIXME: Support multiple, comma-separated, language ranges.
+                simple_selector.pseudo_class.languages.append(pseudo_function.values().first().token().to_string());
             } else if (pseudo_function.name().equals_ignoring_case("nth-child"sv)) {
                 simple_selector.pseudo_class.type = Selector::SimpleSelector::PseudoClass::Type::NthChild;
                 if (!parse_nth_child_pattern(simple_selector, pseudo_function, true))

Userland/Libraries/LibWeb/CSS/Selector.h

@@ -79,6 +79,7 @@ public:
                 Not,
                 Where,
                 Active,
+                Lang,
             };
             Type type { Type::None };
 
@@ -87,6 +88,9 @@ public:
             ANPlusBPattern nth_child_pattern;
 
             SelectorList argument_selector_list {};
+
+            // Used for :lang(en-gb,dk)
+            Vector<FlyString> languages;
         };
         PseudoClass pseudo_class {};
         PseudoElement pseudo_element { PseudoElement::None };
@@ -219,6 +223,8 @@ constexpr StringView pseudo_class_name(Selector::SimpleSelector::PseudoClass::Ty
         return "not"sv;
     case Selector::SimpleSelector::PseudoClass::Type::Where:
         return "where"sv;
+    case Selector::SimpleSelector::PseudoClass::Type::Lang:
+        return "lang"sv;
     case Selector::SimpleSelector::PseudoClass::Type::None:
         break;
     }

Userland/Libraries/LibWeb/CSS/SelectorEngine.cpp

@@ -16,6 +16,34 @@
 
 namespace Web::SelectorEngine {
 
+// https://drafts.csswg.org/selectors-4/#the-lang-pseudo
+static inline bool matches_lang_pseudo_class(DOM::Element const& element, Vector<FlyString> const& languages)
+{
+    FlyString element_language;
+    for (auto const* e = &element; e; e = e->parent_element()) {
+        auto lang = e->attribute(HTML::AttributeNames::lang);
+        if (!lang.is_null()) {
+            element_language = lang;
+            break;
+        }
+    }
+    if (element_language.is_null())
+        return false;
+
+    // FIXME: This is ad-hoc. Implement a proper language range matching algorithm as recommended by BCP47.
+    for (auto const& language : languages) {
+        if (language.is_empty())
+            return false;
+        if (language == "*"sv)
+            return true;
+        if (!element_language.view().contains('-'))
+            return element_language.equals_ignoring_case(language);
+        auto parts = element_language.view().split_view('-');
+        return parts[0].equals_ignoring_case(language);
+    }
+    return false;
+}
+
 static inline bool matches_hover_pseudo_class(DOM::Element const& element)
 {
     auto* hovered_node = element.document().hovered_node();
@@ -140,6 +168,8 @@ static inline bool matches_pseudo_class(CSS::Selector::SimpleSelector::PseudoCla
         return !next_sibling_with_same_tag_name(element);
     case CSS::Selector::SimpleSelector::PseudoClass::Type::OnlyOfType:
         return !previous_sibling_with_same_tag_name(element) && !next_sibling_with_same_tag_name(element);
+    case CSS::Selector::SimpleSelector::PseudoClass::Type::Lang:
+        return matches_lang_pseudo_class(element, pseudo_class.languages);
     case CSS::Selector::SimpleSelector::PseudoClass::Type::Disabled:
         if (!element.tag_name().equals_ignoring_case(HTML::TagNames::input))
             return false;

Userland/Libraries/LibWeb/Dump.cpp

@@ -438,10 +438,17 @@ void dump_selector(StringBuilder& builder, CSS::Selector const& selector)
                 case CSS::Selector::SimpleSelector::PseudoClass::Type::Where:
                     pseudo_class_description = "Where";
                     break;
+                case CSS::Selector::SimpleSelector::PseudoClass::Type::Lang:
+                    pseudo_class_description = "Lang";
+                    break;
                 }
 
                 builder.appendff(" pseudo_class={}", pseudo_class_description);
-                if (pseudo_class.type == CSS::Selector::SimpleSelector::PseudoClass::Type::Not
+                if (pseudo_class.type == CSS::Selector::SimpleSelector::PseudoClass::Type::Lang) {
+                    builder.append('(');
+                    builder.join(',', pseudo_class.languages);
+                    builder.append(')');
+                } else if (pseudo_class.type == CSS::Selector::SimpleSelector::PseudoClass::Type::Not
                     || pseudo_class.type == CSS::Selector::SimpleSelector::PseudoClass::Type::Is
                     || pseudo_class.type == CSS::Selector::SimpleSelector::PseudoClass::Type::Where) {
                     builder.append("([");