Start on infoset coercison

RoboFile.php

@@ -6,6 +6,7 @@ const BASE = __DIR__.\DIRECTORY_SEPARATOR;
 const BASE_TEST = BASE."tests".\DIRECTORY_SEPARATOR;
 define("IS_WIN", defined("PHP_WINDOWS_VERSION_MAJOR"));
 define("IS_MAC", php_uname("s") === "Darwin");
+error_reporting(0);
 
 function norm(string $path): string {
     $out = realpath($path);

lib/DOM/Document.php

@@ -3,7 +3,7 @@ declare(strict_types=1);
 namespace dW\HTML5;
 
 class Document extends \DOMDocument {
-    use Descendant, Serialize;
+    use Descendant, Serialize, EscapeString;
 
     // Quirks mode constants
     public const NO_QUIRKS_MODE = 0;
@@ -11,6 +11,8 @@ class Document extends \DOMDocument {
     public const LIMITED_QUIRKS_MODE = 2;
 
     public $quirksMode = self::NO_QUIRKS_MODE;
+    public $mangledElements = false;
+    public $mangledAttributes = false;
 
     // An array of all template elements created in the document
     // This exists because values of properties on derived DOM classes
@@ -87,21 +89,38 @@ class Document extends \DOMDocument {
     }
 
     public function createElement($name, $value = "") {
-        $e = parent::createElement($name, $value);
-        if ($name === "template") {
-            $this->templateElements[] = $e;
-            $e->content = $this->createDocumentFragment();
+        try {
+            $e = parent::createElement($name, $value);
+            if ($name === "template") {
+                $this->templateElements[] = $e;
+                $e->content = $this->createDocumentFragment();
+            }
+            return $e;
+        } catch (\DOMException) {
+            // The element name is invalid for XML
+            // Replace any offending characters with "UHHHHHH" where H is the
+            //   uppercase hexadecimal digits of the character's code point
+            $this->mangledElements = true;
+            $name = $this->CoerceName($name);
         }
-        return $e;
     }
 
     public function createElementNS($namespaceURI, $qualifiedName, $value = "") {
-        $e = parent::createElementNS($namespaceURI, $qualifiedName, $value);
-        if ($qualifiedName === "template" && $namespaceURI === null) {
-            $this->templateElements[] = $e;
-            $e->content = $this->createDocumentFragment();
+        try {
+            $e = parent::createElementNS($namespaceURI, $qualifiedName, $value);
+            if ($qualifiedName === "template" && $namespaceURI === null) {
+                $this->templateElements[] = $e;
+                $e->content = $this->createDocumentFragment();
+            }
+            return $e;
+        } catch (\DOMException) {
+            throw $e;
+            // The element name is invalid for XML
+            // Replace any offending characters with "UHHHHHH" where H is the
+            //   uppercase hexadecimal digits of the character's code point
+            $this->mangledElements = true;
+            $qualifiedName = $this->CoerceName($qualifiedName);
         }
-        return $e;
     }
 
     public function __toString() {

lib/DOM/traits/EscapeString.php

@@ -22,4 +22,13 @@ trait EscapeString {
 
         return $string;
     }
-}
+
+    protected function CoerceName(string $name): string {
+        // This matches the inverse of the production of NameChar in XML 1.0,
+        //   with the added exclusion of ":" from allowed characters
+        // See https://www.w3.org/TR/REC-xml/#NT-NameStartChar
+        preg_match_all('/[^\-\.0-9\x{B7}\x{300}-\x{36F}\x{203F}-\x{2040}A-Za-z_\x{C0}-\x{D6}\x{D8}-\x{F6}\x{F8}-\x{2FF}\x{370}-\x{37D}\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}]/u', $name, $m, \PREG_OFFSET_CAPTURE);
+        var_export($m);
+        exit;
+    }
+}

lib/TreeBuilder.php

@@ -3347,6 +3347,7 @@ class TreeBuilder {
         $formattingElementIndex = $this->activeFormattingElementsList->findToMarker($subject);
         if ($formattingElementIndex > -1) {
             $formattingElement = $this->activeFormattingElementsList[$formattingElementIndex]['element'];
+            $formattingToken = $this->activeFormattingElementsList[$formattingElementIndex]['token'];
         } else {
             $formattingElement = null;
         }
@@ -3465,8 +3466,9 @@ class TreeBuilder {
         $element = $this->createElementForToken($token, null, $commonAncestor);
         $this->activeFormattingElementsList[$nodeListPos] = ['token' => $nodeToken, 'element' => $element];
         $this->stack[$nodeIndex] = $element;
+        $node = $element;
         # If last node is furthest block, then move the aforementioned
-        #   to be immediately after the new node in the list of
+        #   bookmark to be immediately after the new node in the list of
         #   active formatting elements.
         if ($lastNode->isSameNode($furthestBlock)) {
             $bookmark = $nodeListPos + 1;
@@ -3494,7 +3496,6 @@ class TreeBuilder {
         # Create an element for the token for which formatting element was
         #   created, in the HTML namespace, with furthest block as the
         #   intended parent.
-        $formattingToken = $this->activeFormattingElementsList[$formattingElementIndex]['token'];
         $element = $this->createElementForToken($formattingToken, null, $furthestBlock);
         # Take all of the child nodes of furthest block and append them to
         #   the element created in the last step.
@@ -3511,9 +3512,9 @@ class TreeBuilder {
         # Remove formatting element from the stack of open elements, and
         #   insert the new element into the stack of open elements
         #   immediately below the position of furthest block in that stack.
-        $this->stack->insert($element, $this->stack->findSame($furthestBlock));
         assert($stackIndex > 0, new \Exception("Attempting to delete root element from stack"));
-        unset($this->stack[$this->stack->findSame($formattingElement)]);
+        $this->stack->removeSame($formattingElement);
+        $this->stack->insert($element, $this->stack->findSame($furthestBlock) + 1);
         # Jump back to the step labeled outer loop.
         goto OuterLoop;
     }