LibWeb+Browser: Add ability to run Browser with the new HTML parser

You can now pass "-n" to the browser to use the new HTML parser.
It's not turned on by default since it's still very immature, but this
is a huge step towards bringing it into maturity. :^)

Applications/Browser/Tab.cpp

@@ -59,6 +59,7 @@
 
 namespace Browser {
 
+extern bool g_use_new_html_parser;
 extern String g_home_url;
 
 Tab::Tab()
@@ -70,6 +71,8 @@ Tab::Tab()
     auto& toolbar = m_toolbar_container->add<GUI::ToolBar>();
     m_html_widget = widget.add<Web::HtmlView>();
 
+    m_html_widget->set_use_new_parser(g_use_new_html_parser);
+
     m_go_back_action = GUI::CommonActions::make_go_back_action([this](auto&) {
         m_history.go_back();
         update_actions();

Applications/Browser/main.cpp

@@ -28,6 +28,7 @@
 #include "InspectorWidget.h"
 #include "Tab.h"
 #include "WindowActions.h"
+#include <LibCore/ArgsParser.h>
 #include <LibCore/ConfigFile.h>
 #include <LibCore/File.h>
 #include <LibGUI/AboutDialog.h>
@@ -44,6 +45,7 @@ namespace Browser {
 
 static const char* bookmarks_filename = "/home/anon/bookmarks.json";
 String g_home_url;
+bool g_use_new_html_parser = false;
 
 }
 
@@ -59,6 +61,13 @@ int main(int argc, char** argv)
         return 1;
     }
 
+    const char* specified_url = nullptr;
+
+    Core::ArgsParser args_parser;
+    args_parser.add_option(Browser::g_use_new_html_parser, "Use new HTML parser", "new-parser", 'n');
+    args_parser.add_positional_argument(specified_url, "URL to open", "url", Core::ArgsParser::Required::No);
+    args_parser.parse(argc, argv);
+
     GUI::Application app(argc, argv);
 
     // Connect to the ProtocolServer immediately so we can drop the "unix" pledge.
@@ -174,8 +183,8 @@ int main(int argc, char** argv)
     };
 
     URL first_url = Browser::g_home_url;
-    if (app.args().size() >= 1)
+    if (specified_url)
-        first_url = URL::create_with_url_or_path(app.args()[0]);
+        first_url = URL::create_with_url_or_path(specified_url);
 
     window_actions.on_create_new_tab = [&] {
         create_new_tab(Browser::g_home_url, true);

Libraries/LibWeb/HtmlView.cpp

@@ -47,6 +47,7 @@
 #include <LibWeb/HtmlView.h>
 #include <LibWeb/Layout/LayoutDocument.h>
 #include <LibWeb/Layout/LayoutNode.h>
+#include <LibWeb/Parser/HTMLDocumentParser.h>
 #include <LibWeb/Parser/HTMLParser.h>
 #include <LibWeb/RenderingContext.h>
 #include <LibWeb/ResourceLoader.h>
@@ -431,7 +432,7 @@ static String guess_mime_type_based_on_filename(const URL& url)
     return "text/plain";
 }
 
-static RefPtr<Document> create_document_from_mime_type(const ByteBuffer& data, const URL& url, const String& mime_type, const String& encoding)
+RefPtr<Document> HtmlView::create_document_from_mime_type(const ByteBuffer& data, const URL& url, const String& mime_type, const String& encoding)
 {
     if (mime_type.starts_with("image/"))
         return create_image_document(data, url);
@@ -441,8 +442,14 @@ static RefPtr<Document> create_document_from_mime_type(const ByteBuffer& data, c
         return create_markdown_document(data, url);
     if (mime_type == "text/gemini")
         return create_gemini_document(data, url);
-    if (mime_type == "text/html")
+    if (mime_type == "text/html") {
+        if (m_use_new_parser) {
+            HTMLDocumentParser parser(data);
+            parser.run(url);
+            return parser.document();
+        }
         return parse_html_document(data, url, encoding);
+    }
     return nullptr;
 }
 

Libraries/LibWeb/HtmlView.h

@@ -38,6 +38,9 @@ class HtmlView : public GUI::ScrollableWidget {
 public:
     virtual ~HtmlView() override;
 
+    // FIXME: Remove this once the new parser is ready.
+    void set_use_new_parser(bool use_new_parser) { m_use_new_parser = use_new_parser; }
+
     Document* document();
     const Document* document() const;
     void set_document(Document*);
@@ -83,6 +86,8 @@ protected:
 private:
     virtual void did_scroll() override;
 
+    RefPtr<Document> create_document_from_mime_type(const ByteBuffer& data, const URL& url, const String& mime_type, const String& encoding);
+
     void run_javascript_url(const String& url);
     void layout_and_sync_size();
     void dump_selection(const char* event_name);
@@ -92,6 +97,8 @@ private:
 
     bool m_should_show_line_box_borders { false };
     bool m_in_mouse_selection { false };
+
+    bool m_use_new_parser { false };
 };
 
 }