LibWeb: Implement Attribute closer to the spec and with an IDL file

Note our Attribute class is what the spec refers to as just "Attr". The
main differences between the existing implementation and the spec are
just that the spec defines more fields.

Attributes can contain namespace URIs and prefixes. However, note that
these are not parsed in HTML documents unless the document content-type
is XML. So for now, these are initialized to null. Web pages are able to
set the namespace via JavaScript (setAttributeNS), so these fields may
be filled in when the corresponding APIs are implemented.

The main change to be aware of is that an attribute is a node. This has
implications on how attributes are stored in the Element class. Nodes
are non-copyable and non-movable because these constructors are deleted
by the EventTarget base class. This means attributes cannot be stored in
a Vector or HashMap as these containers assume copyability / movability.
So for now, the Vector holding attributes is changed to hold RefPtrs to
attributes instead. This might change when attribute storage is
implemented according to the spec (by way of NamedNodeMap).

10 files changed, 102 insertions, 42 deletions

diff --git a/Userland/Libraries/LibWeb/CMakeLists.txt b/Userland/Libraries/LibWeb/CMakeLists.txt
index 2f407f7703..f16da893c3 100644
--- a/Userland/Libraries/LibWeb/CMakeLists.txt
+++ b/Userland/Libraries/LibWeb/CMakeLists.txt
@@ -54,6 +54,8 @@ set(SOURCES
     Cookie/ParsedCookie.cpp
     DOM/AbortController.cpp
     DOM/AbortSignal.cpp
+    DOM/Attribute.cpp
+    DOM/Attribute.idl
     DOM/CharacterData.cpp
     DOM/CharacterData.idl
     DOM/Comment.cpp
@@ -366,6 +368,7 @@ libweb_js_wrapper(CSS/MediaQueryListEvent)
 libweb_js_wrapper(CSS/Screen)
 libweb_js_wrapper(CSS/StyleSheet)
 libweb_js_wrapper(CSS/StyleSheetList)
+libweb_js_wrapper(DOM/Attribute)
 libweb_js_wrapper(DOM/AbortController)
 libweb_js_wrapper(DOM/AbortSignal)
 libweb_js_wrapper(DOM/CharacterData)
diff --git a/Userland/Libraries/LibWeb/DOM/Attribute.cpp b/Userland/Libraries/LibWeb/DOM/Attribute.cpp
new file mode 100644
index 0000000000..996a80a404
--- /dev/null
+++ b/Userland/Libraries/LibWeb/DOM/Attribute.cpp
@@ -0,0 +1,24 @@
+/*
+ * Copyright (c) 2021, Tim Flynn <trflynn89@pm.me>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <LibWeb/DOM/Attribute.h>
+#include <LibWeb/DOM/Document.h>
+
+namespace Web::DOM {
+
+NonnullRefPtr<Attribute> Attribute::create(Document& document, FlyString local_name, String value)
+{
+    return adopt_ref(*new Attribute(document, move(local_name), move(value)));
+}
+
+Attribute::Attribute(Document& document, FlyString local_name, String value)
+    : Node(document, NodeType::ATTRIBUTE_NODE)
+    , m_qualified_name(move(local_name), {}, {})
+    , m_value(move(value))
+{
+}
+
+}
diff --git a/Userland/Libraries/LibWeb/DOM/Attribute.h b/Userland/Libraries/LibWeb/DOM/Attribute.h
index 5261a5bf7d..0a9dfd97b9 100644
--- a/Userland/Libraries/LibWeb/DOM/Attribute.h
+++ b/Userland/Libraries/LibWeb/DOM/Attribute.h
@@ -7,25 +7,42 @@
 #pragma once
 
 #include <AK/FlyString.h>
+#include <LibWeb/DOM/Node.h>
+#include <LibWeb/QualifiedName.h>
 
 namespace Web::DOM {
 
-class Attribute {
+// https://dom.spec.whatwg.org/#attr
+class Attribute final : public Node {
 public:
-    Attribute(const FlyString& name, const String& value)
-        : m_name(name)
-        , m_value(value)
-    {
-    }
+    using WrapperType = Bindings::AttributeWrapper;
 
-    const FlyString& name() const { return m_name; }
-    const String& value() const { return m_value; }
+    static NonnullRefPtr<Attribute> create(Document&, FlyString local_name, String value);
 
-    void set_value(const String& value) { m_value = value; }
+    virtual ~Attribute() override = default;
+
+    virtual FlyString node_name() const override { return name(); }
+
+    FlyString const& namespace_uri() const { return m_qualified_name.namespace_(); }
+    FlyString const& prefix() const { return m_qualified_name.prefix(); }
+    FlyString const& local_name() const { return m_qualified_name.local_name(); }
+    String const& name() const { return m_qualified_name.as_string(); }
+
+    String const& value() const { return m_value; }
+    void set_value(String value) { m_value = move(value); }
+
+    Element const* owner_element() const { return m_owner_element; }
+    void set_owner_element(Element const* owner_element) { m_owner_element = owner_element; }
+
+    // Always returns true: https://dom.spec.whatwg.org/#dom-attr-specified
+    constexpr bool specified() const { return true; }
 
 private:
-    FlyString m_name;
+    Attribute(Document&, FlyString local_name, String value);
+
+    QualifiedName m_qualified_name;
     String m_value;
+    Element const* m_owner_element { nullptr };
 };
 
 }
diff --git a/Userland/Libraries/LibWeb/DOM/Attribute.idl b/Userland/Libraries/LibWeb/DOM/Attribute.idl
new file mode 100644
index 0000000000..8b15866a3f
--- /dev/null
+++ b/Userland/Libraries/LibWeb/DOM/Attribute.idl
@@ -0,0 +1,12 @@
+[Exposed=Window]
+interface Attribute : Node {
+    readonly attribute DOMString? namespaceURI;
+    readonly attribute DOMString? prefix;
+    readonly attribute DOMString localName;
+    readonly attribute DOMString name;
+    [CEReactions] attribute DOMString value;
+
+    readonly attribute Element? ownerElement;
+
+    readonly attribute boolean specified;
+};
diff --git a/Userland/Libraries/LibWeb/DOM/Element.cpp b/Userland/Libraries/LibWeb/DOM/Element.cpp
index bed9e07cfe..e6ea910150 100644
--- a/Userland/Libraries/LibWeb/DOM/Element.cpp
+++ b/Userland/Libraries/LibWeb/DOM/Element.cpp
@@ -49,8 +49,8 @@ Element::~Element()
 Attribute* Element::find_attribute(const FlyString& name)
 {
     for (auto& attribute : m_attributes) {
-        if (attribute.name() == name)
-            return &attribute;
+        if (attribute->name() == name)
+            return attribute;
     }
     return nullptr;
 }
@@ -58,8 +58,8 @@ Attribute* Element::find_attribute(const FlyString& name)
 const Attribute* Element::find_attribute(const FlyString& name) const
 {
     for (auto& attribute : m_attributes) {
-        if (attribute.name() == name)
-            return &attribute;
+        if (attribute->name() == name)
+            return attribute;
     }
     return nullptr;
 }
@@ -82,7 +82,7 @@ ExceptionOr<void> Element::set_attribute(const FlyString& name, const String& va
     if (auto* attribute = find_attribute(name))
         attribute->set_value(value);
     else
-        m_attributes.empend(name, value);
+        m_attributes.append(Attribute::create(document(), name, value));
 
     parse_attribute(name, value);
     return {};
@@ -92,7 +92,7 @@ void Element::remove_attribute(const FlyString& name)
 {
     CSS::StyleInvalidator style_invalidator(document());
 
-    m_attributes.remove_first_matching([&](auto& attribute) { return attribute.name() == name; });
+    m_attributes.remove_first_matching([&](auto& attribute) { return attribute->name() == name; });
 }
 
 bool Element::has_class(const FlyString& class_name, CaseSensitivity case_sensitivity) const
diff --git a/Userland/Libraries/LibWeb/DOM/Element.h b/Userland/Libraries/LibWeb/DOM/Element.h
index d23c640280..cc829ff53d 100644
--- a/Userland/Libraries/LibWeb/DOM/Element.h
+++ b/Userland/Libraries/LibWeb/DOM/Element.h
@@ -67,7 +67,7 @@ public:
     void for_each_attribute(Callback callback) const
     {
         for (auto& attribute : m_attributes)
-            callback(attribute.name(), attribute.value());
+            callback(attribute->name(), attribute->value());
     }
 
     bool has_class(const FlyString&, CaseSensitivity = CaseSensitivity::CaseSensitive) const;
@@ -132,7 +132,7 @@ private:
 
     QualifiedName m_qualified_name;
     String m_html_uppercased_qualified_name;
-    Vector<Attribute> m_attributes;
+    Vector<NonnullRefPtr<Attribute>> m_attributes;
 
     RefPtr<CSS::CSSStyleDeclaration> m_inline_style;
 
diff --git a/Userland/Libraries/LibWeb/Forward.h b/Userland/Libraries/LibWeb/Forward.h
index ab862c7c11..2f3dd779dd 100644
--- a/Userland/Libraries/LibWeb/Forward.h
+++ b/Userland/Libraries/LibWeb/Forward.h
@@ -71,6 +71,7 @@ class UnsetStyleValue;
 namespace Web::DOM {
 class AbortController;
 class AbortSignal;
+class Attribute;
 class CharacterData;
 class Comment;
 class CustomEvent;
@@ -290,6 +291,7 @@ class URLSearchParamsIterator;
 namespace Web::Bindings {
 class AbortControllerWrapper;
 class AbortSignalWrapper;
+class AttributeWrapper;
 class CanvasRenderingContext2DWrapper;
 class CharacterDataWrapper;
 class CloseEventWrapper;
diff --git a/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.cpp b/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.cpp
index 460e0a3872..f0c5c524b1 100644
--- a/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.cpp
+++ b/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.cpp
@@ -9,6 +9,8 @@
 #include <AK/StringView.h>
 #include <AK/Utf8View.h>
 #include <LibTextCodec/Decoder.h>
+#include <LibWeb/DOM/Attribute.h>
+#include <LibWeb/DOM/Document.h>
 #include <LibWeb/HTML/Parser/HTMLEncodingDetection.h>
 #include <ctype.h>
 
@@ -94,7 +96,7 @@ Optional<String> extract_character_encoding_from_meta_element(String const& stri
     return TextCodec::get_standardized_encoding(encoding);
 }
 
-Optional<DOM::Attribute> prescan_get_attribute(const ByteBuffer& input, size_t& position)
+RefPtr<DOM::Attribute> prescan_get_attribute(DOM::Document& document, const ByteBuffer& input, size_t& position)
 {
     if (!prescan_skip_whitespace_and_slashes(input, position))
         return {};
@@ -109,7 +111,7 @@ Optional<DOM::Attribute> prescan_get_attribute(const ByteBuffer& input, size_t&
         } else if (input[position] == '\t' || input[position] == '\n' || input[position] == '\f' || input[position] == '\r' || input[position] == ' ')
             goto spaces;
         else if (input[position] == '/' || input[position] == '>')
-            return DOM::Attribute(attribute_name.to_string(), "");
+            return DOM::Attribute::create(document, attribute_name.to_string(), "");
         else
             attribute_name.append_as_lowercase(input[position]);
         ++position;
@@ -121,7 +123,7 @@ spaces:
     if (!prescan_skip_whitespace_and_slashes(input, position))
         return {};
     if (input[position] != '=')
-        return DOM::Attribute(attribute_name.to_string(), "");
+        return DOM::Attribute::create(document, attribute_name.to_string(), "");
     ++position;
 
 value:
@@ -134,13 +136,13 @@ value:
         ++position;
         for (; !prescan_should_abort(input, position); ++position) {
             if (input[position] == quote_character)
-                return DOM::Attribute(attribute_name.to_string(), attribute_value.to_string());
+                return DOM::Attribute::create(document, attribute_name.to_string(), attribute_value.to_string());
             else
                 attribute_value.append_as_lowercase(input[position]);
         }
         return {};
     } else if (input[position] == '>')
-        return DOM::Attribute(attribute_name.to_string(), "");
+        return DOM::Attribute::create(document, attribute_name.to_string(), "");
     else
         attribute_value.append_as_lowercase(input[position]);
 
@@ -150,7 +152,7 @@ value:
 
     for (; !prescan_should_abort(input, position); ++position) {
         if (input[position] == '\t' || input[position] == '\n' || input[position] == '\f' || input[position] == '\r' || input[position] == ' ' || input[position] == '>')
-            return DOM::Attribute(attribute_name.to_string(), attribute_value.to_string());
+            return DOM::Attribute::create(document, attribute_name.to_string(), attribute_value.to_string());
         else
             attribute_value.append_as_lowercase(input[position]);
     }
@@ -158,7 +160,7 @@ value:
 }
 
 // https://html.spec.whatwg.org/multipage/parsing.html#prescan-a-byte-stream-to-determine-its-encoding
-Optional<String> run_prescan_byte_stream_algorithm(const ByteBuffer& input)
+Optional<String> run_prescan_byte_stream_algorithm(DOM::Document& document, const ByteBuffer& input)
 {
     // https://html.spec.whatwg.org/multipage/parsing.html#prescan-a-byte-stream-to-determine-its-encoding
 
@@ -194,24 +196,24 @@ Optional<String> run_prescan_byte_stream_algorithm(const ByteBuffer& input)
             Optional<String> charset {};
 
             while (true) {
-                auto attribute = prescan_get_attribute(input, position);
-                if (!attribute.has_value())
+                auto attribute = prescan_get_attribute(document, input, position);
+                if (!attribute)
                     break;
-                if (attribute_list.contains_slow(attribute.value().name()))
+                if (attribute_list.contains_slow(attribute->name()))
                     continue;
-                auto& attribute_name = attribute.value().name();
-                attribute_list.append(attribute.value().name());
+                auto& attribute_name = attribute->name();
+                attribute_list.append(attribute->name());
 
                 if (attribute_name == "http-equiv") {
-                    got_pragma = attribute.value().value() == "content-type";
+                    got_pragma = attribute->value() == "content-type";
                 } else if (attribute_name == "content") {
-                    auto encoding = extract_character_encoding_from_meta_element(attribute.value().value());
+                    auto encoding = extract_character_encoding_from_meta_element(attribute->value());
                     if (encoding.has_value() && !charset.has_value()) {
                         charset = encoding.value();
                         need_pragma = true;
                     }
                 } else if (attribute_name == "charset") {
-                    auto maybe_charset = TextCodec::get_standardized_encoding(attribute.value().value());
+                    auto maybe_charset = TextCodec::get_standardized_encoding(attribute->value());
                     if (maybe_charset.has_value()) {
                         charset = Optional<String> { maybe_charset };
                         need_pragma = { false };
@@ -231,7 +233,7 @@ Optional<String> run_prescan_byte_stream_algorithm(const ByteBuffer& input)
             && ((input[position + 1] == '/' && isalpha(input[position + 2])) || isalpha(input[position + 1]))) {
             position += 2;
             prescan_skip_whitespace_and_slashes(input, position);
-            while (prescan_get_attribute(input, position).has_value()) { };
+            while (prescan_get_attribute(document, input, position)) { };
         } else if (!prescan_should_abort(input, position + 1) && input[position] == '<' && (input[position + 1] == '!' || input[position + 1] == '/' || input[position + 1] == '?')) {
             position += 2;
             while (input[position] != '>') {
@@ -247,7 +249,7 @@ Optional<String> run_prescan_byte_stream_algorithm(const ByteBuffer& input)
 }
 
 // https://html.spec.whatwg.org/multipage/parsing.html#determining-the-character-encoding
-String run_encoding_sniffing_algorithm(const ByteBuffer& input)
+String run_encoding_sniffing_algorithm(DOM::Document& document, const ByteBuffer& input)
 {
     if (input.size() >= 2) {
         if (input[0] == 0xFE && input[1] == 0xFF) {
@@ -265,7 +267,7 @@ String run_encoding_sniffing_algorithm(const ByteBuffer& input)
     //        at any later step in this algorithm.
     // FIXME: If the transport layer specifies a character encoding, and it is supported.
 
-    auto optional_encoding = run_prescan_byte_stream_algorithm(input);
+    auto optional_encoding = run_prescan_byte_stream_algorithm(document, input);
     if (optional_encoding.has_value()) {
         return optional_encoding.value();
     }
diff --git a/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.h b/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.h
index 4d9a1e9ab2..52784e8d7c 100644
--- a/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.h
+++ b/Userland/Libraries/LibWeb/HTML/Parser/HTMLEncodingDetection.h
@@ -8,7 +8,7 @@
 
 #include <AK/Optional.h>
 #include <AK/String.h>
-#include <LibWeb/DOM/Attribute.h>
+#include <LibWeb/Forward.h>
 
 namespace Web::HTML {
 
@@ -16,8 +16,8 @@ bool prescan_should_abort(const ByteBuffer& input, const size_t& position);
 bool prescan_is_whitespace_or_slash(const u8& byte);
 bool prescan_skip_whitespace_and_slashes(const ByteBuffer& input, size_t& position);
 Optional<String> extract_character_encoding_from_meta_element(String const&);
-Optional<DOM::Attribute> prescan_get_attribute(const ByteBuffer& input, size_t& position);
-Optional<String> run_prescan_byte_stream_algorithm(const ByteBuffer& input);
-String run_encoding_sniffing_algorithm(const ByteBuffer& input);
+RefPtr<DOM::Attribute> prescan_get_attribute(DOM::Document&, const ByteBuffer& input, size_t& position);
+Optional<String> run_prescan_byte_stream_algorithm(DOM::Document&, const ByteBuffer& input);
+String run_encoding_sniffing_algorithm(DOM::Document&, const ByteBuffer& input);
 
 }
diff --git a/Userland/Libraries/LibWeb/HTML/Parser/HTMLParser.cpp b/Userland/Libraries/LibWeb/HTML/Parser/HTMLParser.cpp
index d9fc29f64b..1a7ada5a10 100644
--- a/Userland/Libraries/LibWeb/HTML/Parser/HTMLParser.cpp
+++ b/Userland/Libraries/LibWeb/HTML/Parser/HTMLParser.cpp
@@ -2810,7 +2810,7 @@ void HTMLParser::handle_in_frameset(HTMLToken& token)
     }
 
     if (token.is_end_of_file()) {
-        //FIXME: If the current node is not the root html element, then this is a parse error.
+        // FIXME: If the current node is not the root html element, then this is a parse error.
 
         stop_parsing();
         return;
@@ -3162,7 +3162,7 @@ NonnullOwnPtr<HTMLParser> HTMLParser::create_with_uncertain_encoding(DOM::Docume
 {
     if (document.has_encoding())
         return make<HTMLParser>(document, input, document.encoding().value());
-    auto encoding = run_encoding_sniffing_algorithm(input);
+    auto encoding = run_encoding_sniffing_algorithm(document, input);
     dbgln("The encoding sniffing algorithm returned encoding '{}'", encoding);
     return make<HTMLParser>(document, input, encoding);
 }