LibHTML: Add Comment and CharacterData nodes and improve HTML parsing

This patch adds the CharacterData subclass of Node, which is now the
parent class of Text and a new Comment class.

A Comment node is one of these in HTML: <!--hello friends-->
Since these occur somewhat frequently on the web, we need to be able
to parse them.

This patch also adds a child rejection mechanism to the DOM tree.
Nodes can now override is_child_allowed(Node) and return false if they
don't want a particular Node to become a child of theirs. This is used
to prevent Document from taking on unwanted children.

Libraries/LibHTML/DOM/CharacterData.cpp

@@ -0,0 +1,11 @@
+#include <LibHTML/DOM/CharacterData.h>
+
+CharacterData::CharacterData(Document& document, NodeType type, const String& data)
+    : Node(document, type)
+    , m_data(data)
+{
+}
+
+CharacterData::~CharacterData()
+{
+}

Libraries/LibHTML/DOM/CharacterData.h

@@ -0,0 +1,25 @@
+#pragma once
+
+#include <AK/String.h>
+#include <LibHTML/DOM/Node.h>
+
+class CharacterData : public Node {
+public:
+    virtual ~CharacterData() override;
+
+    const String& data() const { return m_data; }
+
+    virtual String text_content() const override { return m_data; }
+
+protected:
+    explicit CharacterData(Document&, NodeType, const String&);
+
+private:
+    String m_data;
+};
+
+template<>
+inline bool is<CharacterData>(const Node& node)
+{
+    return node.is_character_data();
+}

Libraries/LibHTML/DOM/Comment.cpp

@@ -0,0 +1,11 @@
+#include <LibHTML/DOM/Comment.h>
+#include <LibHTML/Layout/LayoutText.h>
+
+Comment::Comment(Document& document, const String& data)
+    : CharacterData(document, NodeType::COMMENT_NODE, data)
+{
+}
+
+Comment::~Comment()
+{
+}

Libraries/LibHTML/DOM/Comment.h

@@ -0,0 +1,18 @@
+#pragma once
+
+#include <AK/String.h>
+#include <LibHTML/DOM/CharacterData.h>
+
+class Comment final : public CharacterData {
+public:
+    explicit Comment(Document&, const String&);
+    virtual ~Comment() override;
+
+    virtual String tag_name() const override { return "#comment"; }
+};
+
+template<>
+inline bool is<Comment>(const Node& node)
+{
+    return node.is_comment();
+}

Libraries/LibHTML/DOM/Document.cpp

@@ -29,6 +29,23 @@ StyleResolver& Document::style_resolver()
     return *m_style_resolver;
 }
 
+bool Document::is_child_allowed(const Node& node) const
+{
+    switch (node.type()) {
+    case NodeType::DOCUMENT_NODE:
+    case NodeType::TEXT_NODE:
+        return false;
+    case NodeType::COMMENT_NODE:
+        return true;
+    case NodeType::DOCUMENT_TYPE_NODE:
+        return !first_child_of_type<DocumentType>();
+    case NodeType::ELEMENT_NODE:
+        return !first_child_of_type<Element>();
+    default:
+        return false;
+    }
+}
+
 void Document::fixup()
 {
     if (!is<DocumentType>(first_child()))

Libraries/LibHTML/DOM/Document.h

@@ -67,6 +67,8 @@ public:
     void invalidate_layout();
     Function<void()> on_invalidate_layout;
 
+    virtual bool is_child_allowed(const Node&) const override;
+
 private:
     virtual RefPtr<LayoutNode> create_layout_node(const StyleResolver&, const StyleProperties* parent_style) const override;
 

Libraries/LibHTML/DOM/DocumentType.h

@@ -7,7 +7,7 @@ public:
     explicit DocumentType(Document&);
     virtual ~DocumentType() override;
 
-    virtual String tag_name() const override { return "!DOCTYPE"; }
+    virtual String tag_name() const override { return "#doctype"; }
 };
 
 template<>

Libraries/LibHTML/DOM/Node.h

@@ -10,6 +10,7 @@ enum class NodeType : unsigned {
     INVALID = 0,
     ELEMENT_NODE = 1,
     TEXT_NODE = 3,
+    COMMENT_NODE = 8,
     DOCUMENT_NODE = 9,
     DOCUMENT_TYPE_NODE = 10,
 };
@@ -32,6 +33,8 @@ public:
     bool is_text() const { return type() == NodeType::TEXT_NODE; }
     bool is_document() const { return type() == NodeType::DOCUMENT_NODE; }
     bool is_document_type() const { return type() == NodeType::DOCUMENT_TYPE_NODE; }
+    bool is_comment() const { return type() == NodeType::COMMENT_NODE; }
+    bool is_character_data() const { return type() == NodeType::TEXT_NODE || type() == NodeType::COMMENT_NODE; }
     bool is_parent_node() const { return is_element() || is_document(); }
 
     virtual RefPtr<LayoutNode> create_layout_node(const StyleResolver&, const StyleProperties* parent_style) const;
@@ -66,6 +69,8 @@ public:
     const Element* previous_element_sibling() const;
     const Element* next_element_sibling() const;
 
+    virtual bool is_child_allowed(const Node&) const { return true; }
+
 protected:
     Node(Document&, NodeType);
 

Libraries/LibHTML/DOM/Text.cpp

@@ -2,8 +2,7 @@
 #include <LibHTML/Layout/LayoutText.h>
 
 Text::Text(Document& document, const String& data)
-    : Node(document, NodeType::TEXT_NODE)
-    , m_data(data)
+    : CharacterData(document, NodeType::TEXT_NODE, data)
 {
 }
 

Libraries/LibHTML/DOM/Text.h

@@ -1,23 +1,17 @@
 #pragma once
 
 #include <AK/String.h>
-#include <LibHTML/DOM/Node.h>
+#include <LibHTML/DOM/CharacterData.h>
 
-class Text final : public Node {
+class Text final : public CharacterData {
 public:
     explicit Text(Document&, const String&);
     virtual ~Text() override;
 
-    const String& data() const { return m_data; }
-
     virtual String tag_name() const override { return "#text"; }
 
-    virtual String text_content() const override { return m_data; }
-
 private:
     virtual RefPtr<LayoutNode> create_layout_node(const StyleResolver&, const StyleProperties* parent_style) const override;
-
-    String m_data;
 };
 
 template<>