Added MathML Presentation element checking.

2 files changed, 279 insertions, 114 deletions

diff --git a/src/HTML5/Elements.php b/src/HTML5/Elements.php
index d437960..ad58363 100644
--- a/src/HTML5/Elements.php
+++ b/src/HTML5/Elements.php
@@ -1,6 +1,13 @@
 <?php
 namespace HTML5;
 
+/**
+ * Provide general element functions.
+ *
+ * @todo consider using a bitmask table lookup. There is enought overlap in
+ *   naming that this could significantly shrink the size and maybe make it
+ *   faster. See the Go teams implementation at https://code.google.com/p/go/source/browse/html/atom.
+ */
 class Elements {
 
   const TEXT_RAW = 0x01;
@@ -126,6 +133,56 @@ class Elements {
   );
 
   /**
+   * The MathML elements. See http://www.w3.org/wiki/MathML/Elements.
+   *
+   * In our case we are only concerned with presentation MathML and not content
+   * MathML. There is a nice list of this subset at https://developer.mozilla.org/en-US/docs/MathML/Element.
+   * 
+   * @var array
+   */
+  public static $mathml = array(
+    "maction" => 1,
+    "maligngroup" => 1,
+    "malignmark" => 1,
+    "math" => 1,
+    "menclose" => 1,
+    "merror" => 1,
+    "mfenced" => 1,
+    "mfrac" => 1,
+    "mglyph" => 1,
+    "mi" => 1,
+    "mlabeledtr" => 1,
+    "mlongdiv" => 1,
+    "mmultiscripts" => 1,
+    "mn" => 1,
+    "mo" => 1,
+    "mover" => 1,
+    "mpadded" => 1,
+    "mphantom" => 1,
+    "mroot" => 1,
+    "mrow" => 1,
+    "ms" => 1,
+    "mscarries" => 1,
+    "mscarry" => 1,
+    "msgroup" => 1,
+    "msline" => 1,
+    "mspace" => 1,
+    "msqrt" => 1,
+    "msrow" => 1,
+    "mstack" => 1,
+    "mstyle" => 1,
+    "msub" => 1,
+    "msup" => 1,
+    "msubsup" => 1,
+    "mtable" => 1,
+    "mtd" => 1,
+    "mtext" => 1,
+    "mtr" => 1,
+    "munder" => 1,
+    "munderover" => 1,
+  );
+
+  /**
    * Test if an element is a valid html5 element.
    *
    * @param string $name
@@ -140,4 +197,35 @@ class Elements {
     // Do we need this check or will all data passed here already be lowercase?
     return isset(self::$elements[strtolower($name)]);
   }
+
+  /**
+   * Test if an element name is a valid MathML presentation element.
+   *
+   * @param string $name
+   *   The name of the element.
+   *
+   * @return bool
+   *   True if a MathML name and false otherwise.
+   */
+  public static function isMathMLElement($name) {
+
+    // MathML is case-sensetitive unlike html5 elements.
+    return isset(self::$mathml[$name]);
+  }
+
+  /**
+   * Is an element name valid in an html5 document.
+   *
+   * This includes html5 elements along with other allowed embedded content
+   * such as svg and mathml.
+   * 
+   * @param string $name 
+   *   The name of the element.
+   *
+   * @return bool
+   *   True if valid and false otherwise.
+   */
+  public function isElement($name) {
+    return self::isHtml5Element($name) || self::isMathMLElement($name);
+  }
 }
diff --git a/test/HTML5/ElementsTest.php b/test/HTML5/ElementsTest.php
index d3ba43b..318b210 100644
--- a/test/HTML5/ElementsTest.php
+++ b/test/HTML5/ElementsTest.php
@@ -7,121 +7,164 @@ require_once 'TestCase.php';
 
 class ElementsTest extends TestCase {
 
+  public $html5Elements = array(
+    "a",
+    "abbr",
+    "address",
+    "area",
+    "article",
+    "aside",
+    "audio",
+    "b",
+    "base",
+    "bdi",
+    "bdo",
+    "blockquote",
+    "body",
+    "br",
+    "button",
+    "canvas",
+    "caption",
+    "cite",
+    "code",
+    "col",
+    "colgroup",
+    "command",
+    //"data",
+    "datalist",
+    "dd",
+    "del",
+    "details",
+    "dfn",
+    "dialog",
+    "div",
+    "dl",
+    "dt",
+    "em",
+    "embed",
+    "fieldset",
+    "figcaption",
+    "figure",
+    "footer",
+    "form",
+    "h1",
+    "h2",
+    "h3",
+    "h4",
+    "h5",
+    "h6",
+    "head",
+    "header",
+    "hgroup",
+    "hr",
+    "html",
+    "i",
+    "iframe",
+    "img",
+    "input",
+    "ins",
+    "kbd",
+    "keygen",
+    "label",
+    "legend",
+    "li",
+    "link",
+    "map",
+    "mark",
+    "menu",
+    "meta",
+    "meter",
+    "nav",
+    "noscript",
+    "object",
+    "ol",
+    "optgroup",
+    "option",
+    "output",
+    "p",
+    "param",
+    "pre",
+    "progress",
+    "q",
+    "rp",
+    "rt",
+    "ruby",
+    "s",
+    "samp",
+    "script",
+    "section",
+    "select",
+    "small",
+    "source",
+    "span",
+    "strong",
+    "style",
+    "sub",
+    "summary",
+    "sup",
+    "table",
+    "tbody",
+    "td",
+    "textarea",
+    "tfoot",
+    "th",
+    "thead",
+    "time",
+    "title",
+    "tr",
+    "track",
+    "u",
+    "ul",
+    "var",
+    "video",
+    "wbr",
+  );
+
+  public $mathmlElements = array(
+    "maction",
+    "maligngroup",
+    "malignmark",
+    "math",
+    "menclose",
+    "merror",
+    "mfenced",
+    "mfrac",
+    "mglyph",
+    "mi",
+    "mlabeledtr",
+    "mlongdiv",
+    "mmultiscripts",
+    "mn",
+    "mo",
+    "mover",
+    "mpadded",
+    "mphantom",
+    "mroot",
+    "mrow",
+    "ms",
+    "mscarries",
+    "mscarry",
+    "msgroup",
+    "msline",
+    "mspace",
+    "msqrt",
+    "msrow",
+    "mstack",
+    "mstyle",
+    "msub",
+    "msup",
+    "msubsup",
+    "mtable",
+    "mtd",
+    "mtext",
+    "mtr",
+    "munder",
+    "munderover",
+  );
+
   public function testIsHtml5Element() {
-    $html5 = array(
-      "a",
-      "abbr",
-      "address",
-      "area",
-      "article",
-      "aside",
-      "audio",
-      "b",
-      "base",
-      "bdi",
-      "bdo",
-      "blockquote",
-      "body",
-      "br",
-      "button",
-      "canvas",
-      "caption",
-      "cite",
-      "code",
-      "col",
-      "colgroup",
-      "command",
-      //"data",
-      "datalist",
-      "dd",
-      "del",
-      "details",
-      "dfn",
-      "dialog",
-      "div",
-      "dl",
-      "dt",
-      "em",
-      "embed",
-      "fieldset",
-      "figcaption",
-      "figure",
-      "footer",
-      "form",
-      "h1",
-      "h2",
-      "h3",
-      "h4",
-      "h5",
-      "h6",
-      "head",
-      "header",
-      "hgroup",
-      "hr",
-      "html",
-      "i",
-      "iframe",
-      "img",
-      "input",
-      "ins",
-      "kbd",
-      "keygen",
-      "label",
-      "legend",
-      "li",
-      "link",
-      "map",
-      "mark",
-      "menu",
-      "meta",
-      "meter",
-      "nav",
-      "noscript",
-      "object",
-      "ol",
-      "optgroup",
-      "option",
-      "output",
-      "p",
-      "param",
-      "pre",
-      "progress",
-      "q",
-      "rp",
-      "rt",
-      "ruby",
-      "s",
-      "samp",
-      "script",
-      "section",
-      "select",
-      "small",
-      "source",
-      "span",
-      "strong",
-      "style",
-      "sub",
-      "summary",
-      "sup",
-      "table",
-      "tbody",
-      "td",
-      "textarea",
-      "tfoot",
-      "th",
-      "thead",
-      "time",
-      "title",
-      "tr",
-      "track",
-      "u",
-      "ul",
-      "var",
-      "video",
-      "wbr",
-    );
-
-    foreach ($html5 as $element) {
+    
+    foreach ($this->html5Elements as $element) {
       $this->assertTrue(Elements::isHtml5Element($element), 'html5 element test failed on: ' . $element);
 
       $this->assertTrue(Elements::isHtml5Element(strtoupper($element)), 'html5 element test failed on: ' . strtoupper($element));
@@ -133,8 +176,42 @@ class ElementsTest extends TestCase {
 
       $this->assertFalse(Elements::isHtml5Element(strtoupper($element)), 'html5 element test failed on: ' . strtoupper($element));
     }
+  }
+
+  public function testIsMathMLElement() {
+    foreach ($this->mathmlElements as $element) {
+      $this->assertTrue(Elements::isMathMLElement($element), 'MathML element test failed on: ' . $element);
+
+      // MathML is case sensetitive so these should all fail.
+      $this->assertFalse(Elements::isMathMLElement(strtoupper($element)), 'MathML element test failed on: ' . strtoupper($element));
+    }
+
+    $nonMathML = array('foo', 'bar', 'baz');
+    foreach ($nonMathML as $element) {
+      $this->assertFalse(Elements::isMathMLElement($element), 'MathML element test failed on: ' . $element);
+    }
+  }
+
+  public function testIsElement() {
+    foreach ($this->html5Elements as $element) {
+      $this->assertTrue(Elements::isElement($element), 'html5 element test failed on: ' . $element);
+
+      $this->assertTrue(Elements::isElement(strtoupper($element)), 'html5 element test failed on: ' . strtoupper($element));
+    }
 
+    foreach ($this->mathmlElements as $element) {
+      $this->assertTrue(Elements::isElement($element), 'MathML element test failed on: ' . $element);
+
+      // MathML is case sensetitive so these should all fail.
+      $this->assertFalse(Elements::isElement(strtoupper($element)), 'MathML element test failed on: ' . strtoupper($element));
+    }
 
+    $nonhtml5 = array('foo', 'bar', 'baz');
+    foreach ($nonhtml5 as $element) {
+      $this->assertFalse(Elements::isElement($element), 'html5 element test failed on: ' . $element);
+
+      $this->assertFalse(Elements::isElement(strtoupper($element)), 'html5 element test failed on: ' . strtoupper($element));
+    }
   }
 
 }
\ No newline at end of file