More sanitization progress

Script to scrape attribute list needs more work to deal with globals

RoboFile.php

@@ -160,28 +160,54 @@ class RoboFile extends \Robo\Tasks {
     public function constants(): Result {
         $c = $this->collectionBuilder()->addCode(function() {
             $elems = [];
+            $atrs = [];
             // retrieve the single-page HTML specification (this is around 15MB in size)
             $spec = file_get_contents("https://html.spec.whatwg.org/");
             // parse it (this may take several seconds
             if ($spec) {
                 $p = new DOMParser;
                 $document = $p->parseFromString($spec, "text/html;charset=utf-8");
-                // pick out element definitions from the specification
+                // pick out element and attribute definitions from the specification
                 foreach ($document->getElementsByTagName("dfn") as $el) {
-                    if ($el->getAttribute("data-dfn-type") === "element") {
+                    $type = $el->getAttribute("data-dfn-type");
+                    if ($type === "element") {
                         $elems[] = trim($el->textContent);
+                    } elseif ($type === "element-attr") {
+                        $name = trim($el->textContent);
+                        if (preg_match('/\s/', $name)) {
+                            // skip the definition if the name is not valid
+                            continue;
+                        }
+                        $attrs[$name] = $attrs[$name] ?? [];
+                        $context = $el->getAttribute("data-dfn-for") ?? "";
+                        if (strlen($context)) {
+                            $context = explode(",", $context);
+                            $attrs[$name] = array_merge($attrs[$name], $context);
+                        }
                     }
                 }
             }
             // sort and filter the results for unqiueness
             sort($elems);
+            ksort($attrs);
             $elems = array_unique($elems);
-            // output the list of elements as a PHP array
+            $attrs = array_map(function($v) {
+                sort($v);
+                return $v;
+            }, $attrs);
+            // formt the lists as a PHP arrays
             $elems = array_map(function($e) {
                 return "'$e' => true";
             }, $elems);
             $elems = implode(", ", $elems);
+            $attrList = [];
+            foreach ($attrs as $name => $context) {
+                $attrList[] = "'$name' => [\"".implode("\", \"", $context)."\"]";
+            }
+            $attrs = implode(", ", $attrList);
             echo "protected const KNOWN_ELEMENTS_HTML = [".$elems."];\n";
+            echo "protected const KNOWN_ATTRIBUTES_HTML = [".$attrs."];\n";
+            
         });
         return $c->run();
     }