Count element name validity per DOM versus HTML parsing.

third_party/WebKit/Source/core/dom/Document.cpp

 /*
  * Copyright (C) 1999 Lars Knoll (knoll@kde.org)
  *           (C) 1999 Antti Koivisto (koivisto@kde.org)
  *           (C) 2001 Dirk Mueller (mueller@kde.org)
  *           (C) 2006 Alexey Proskuryakov (ap@webkit.org)
  * Copyright (C) 2004, 2005, 2006, 2007, 2008, 2009, 2011, 2012 Apple Inc. All
  * rights reserved.
  * Copyright (C) 2008, 2009 Torch Mobile Inc. All rights reserved.
  * (http://www.torchmobile.com/)
  * Copyright (C) 2008, 2009, 2011, 2012 Google Inc. All rights reserved.
@@ skipping 357 matching lines @@
   // rule (d) above
   WTF::Unicode::CharDecompositionType decomp_type =
       WTF::Unicode::DecompositionType(c);
   if (decomp_type == WTF::Unicode::kDecompositionFont ||
       decomp_type == WTF::Unicode::kDecompositionCompat)
     return false;
 
   return true;
 }
 
+// Tests whether |name| is something the HTML parser would accept as a
+// tag name.
+template <typename CharType>
+static inline bool IsValidElementNamePerHTMLParser(const CharType* characters,
+                                                   unsigned length) {
+  CharType c = characters[0] | 0x20;
+  if (!('a' <= c && c < 'z'))
+    return false;
+
+  for (unsigned i = 1; i < length; ++i) {
+    c = characters[i];
+    if (c == '\t' || c == '\n' || c == '\f' || c == '\r' || c == ' ' ||
+        c == '/' || c == '>')
+      return false;
+  }
+
+  return true;
+}
+
+static bool IsValidElementNamePerHTMLParser(const String& name) {
+  unsigned length = name.length();
+  if (!length)
+    return false;
+
+  if (name.Is8Bit()) {
+    const LChar* characters = name.Characters8();
+    return IsValidElementNamePerHTMLParser(characters, length);
+  }
+  const UChar* characters = name.Characters16();
+  return IsValidElementNamePerHTMLParser(characters, length);
+}
+
+// Tests whether |name| is a valid name per DOM spec. Also checks
+// whether the HTML parser would accept this element name and counts
+// cases of mismatches.
+static bool IsValidElementName(const LocalDOMWindow* window,
+                               const String& name) {
+  bool is_valid_dom_name = Document::IsValidName(name);
+  bool is_valid_html_name = IsValidElementNamePerHTMLParser(name);
+  if (UNLIKELY(is_valid_html_name != is_valid_dom_name && window)) {
+    UseCounter::Count(window->GetFrame(),
+                      is_valid_dom_name
+                          ? UseCounter::kElementNameDOMValidHTMLParserInvalid
+                          : UseCounter::kElementNameDOMInvalidHTMLParserValid);
+  }
+  return is_valid_dom_name;
+}
+
 static bool AcceptsEditingFocus(const Element& element) {
   DCHECK(HasEditableStyle(element));
 
   return element.GetDocument().GetFrame() && RootEditableElement(element);
 }
 
 uint64_t Document::global_tree_version_ = 0;
 
 static bool g_threaded_parsing_enabled_for_testing = true;
 
@@ skipping 299 matching lines @@
 
 bool Document::IsInMainFrame() const {
   return GetFrame() && GetFrame()->IsMainFrame();
 }
 
 AtomicString Document::ConvertLocalName(const AtomicString& name) {
   return IsHTMLDocument() ? name.LowerASCII() : name;
 }
 
 // https://dom.spec.whatwg.org/#dom-document-createelement
-Element* Document::createElement(const AtomicString& name,
+Element* Document::createElement(const LocalDOMWindow* window,
+                                 const AtomicString& name,
                                  ExceptionState& exception_state) {
-  if (!IsValidName(name)) {
+  if (!IsValidElementName(window, name)) {
     exception_state.ThrowDOMException(
         kInvalidCharacterError,
         "The tag name provided ('" + name + "') is not a valid name.");
     return nullptr;
   }
 
   if (IsXHTMLDocument() || IsHTMLDocument()) {
     // 2. If the context object is an HTML document, let localName be
     // converted to ASCII lowercase.
     AtomicString local_name = ConvertLocalName(name);
@@ skipping 29 matching lines @@
       return g_empty_string;
 
     if (impl.hasIs())
       return impl.is();
   }
 
   return g_empty_string;
 }
 
 // https://dom.spec.whatwg.org/#dom-document-createelement
-Element* Document::createElement(const AtomicString& local_name,
+Element* Document::createElement(const LocalDOMWindow* window,
+                                 const AtomicString& local_name,
                                  const StringOrDictionary& string_or_options,
                                  ExceptionState& exception_state) {
   // 1. If localName does not match Name production, throw InvalidCharacterError
-  if (!IsValidName(local_name)) {
+  if (!IsValidElementName(window, local_name)) {
     exception_state.ThrowDOMException(
         kInvalidCharacterError,
         "The tag name provided ('" + local_name + "') is not a valid name.");
     return nullptr;
   }
 
   // 2. localName converted to ASCII lowercase
   const AtomicString& converted_local_name = ConvertLocalName(local_name);
 
   bool is_v1 = string_or_options.isDictionary() || !RegistrationContext();
@@ skipping 35 matching lines @@
 
   if (definition) {
     element = CustomElement::CreateCustomElementSync(
         *this, converted_local_name, definition);
   } else if (V0CustomElement::IsValidName(local_name) &&
              RegistrationContext()) {
     element = RegistrationContext()->CreateCustomTagElement(
         *this,
         QualifiedName(g_null_atom, converted_local_name, xhtmlNamespaceURI));
   } else {
-    element = createElement(local_name, exception_state);
+    element = createElement(window, local_name, exception_state);
     if (exception_state.HadException())
       return nullptr;
   }
 
   // 8. If 'is' is non-null, set 'is' attribute
   if (!is.IsEmpty()) {
     if (string_or_options.isString()) {
       V0CustomElementRegistrationContext::SetIsAttributeAndTypeExtension(
           element, is);
     } else if (string_or_options.isDictionary()) {
@@ skipping 19 matching lines @@
         kNamespaceError,
         "The namespace URI provided ('" + namespace_uri +
             "') is not valid for the qualified name provided ('" +
             qualified_name + "').");
     return QualifiedName::Null();
   }
 
   return q_name;
 }
 
-Element* Document::createElementNS(const AtomicString& namespace_uri,
+Element* Document::createElementNS(const LocalDOMWindow* window,
+                                   const AtomicString& namespace_uri,
                                    const AtomicString& qualified_name,
                                    ExceptionState& exception_state) {
   QualifiedName q_name(
       CreateQualifiedName(namespace_uri, qualified_name, exception_state));
   if (q_name == QualifiedName::Null())
     return nullptr;
 
   if (CustomElement::ShouldCreateCustomElement(q_name))
     return CustomElement::CreateCustomElementSync(*this, q_name);
   return createElement(q_name, kCreatedByCreateElement);
 }
 
 // https://dom.spec.whatwg.org/#internal-createelementns-steps
-Element* Document::createElementNS(const AtomicString& namespace_uri,
+Element* Document::createElementNS(const LocalDOMWindow* window,
+                                   const AtomicString& namespace_uri,
                                    const AtomicString& qualified_name,
                                    const StringOrDictionary& string_or_options,
                                    ExceptionState& exception_state) {
   // 1. Validate and extract
   QualifiedName q_name(
       CreateQualifiedName(namespace_uri, qualified_name, exception_state));
   if (q_name == QualifiedName::Null())
     return nullptr;
 
   bool is_v1 = string_or_options.isDictionary() || !RegistrationContext();
   bool create_v1_builtin =
       string_or_options.isDictionary() &&
       RuntimeEnabledFeatures::customElementsBuiltinEnabled();
   bool should_create_builtin =
       create_v1_builtin || string_or_options.isString();
 
   // 2.
   const AtomicString& is =
       AtomicString(GetTypeExtension(this, string_or_options, exception_state));
   const AtomicString& name = should_create_builtin ? is : qualified_name;
 
-  if (!IsValidName(qualified_name)) {
+  if (!IsValidElementName(window, qualified_name)) {
     exception_state.ThrowDOMException(
         kInvalidCharacterError, "The tag name provided ('" + qualified_name +
                                     "') is not a valid name.");
     return nullptr;
   }
 
   // 3. Let definition be result of lookup up custom element definition
   CustomElementDefinition* definition = nullptr;
   if (is_v1) {
     const CustomElementDescriptor desc =
@@ skipping 5795 matching lines @@
 }
 
 void showLiveDocumentInstances() {
   WeakDocumentSet& set = liveDocumentSet();
   fprintf(stderr, "There are %u documents currently alive:\n", set.size());
   for (blink::Document* document : set)
     fprintf(stderr, "- Document %p URL: %s\n", document,
             document->Url().GetString().Utf8().data());
 }
 #endif