[FEATURE] Add HtmlPruner::removeRedundantClasses

CHANGELOG.md

@@ -6,6 +6,9 @@ This project adheres to [Semantic Versioning](https://semver.org/).
 ## x.y.z
 
 ### Added
+- Add `HtmlPruner::removeRedundantClasses`
+  ([#380](https://github.com/MyIntervals/emogrifier/issues/380),
+  [#708](https://github.com/MyIntervals/emogrifier/pull/708))
 - Support multiple attributes selectors
   ([#385](https://github.com/MyIntervals/emogrifier/issues/385),
   [#721](https://github.com/MyIntervals/emogrifier/pull/721))

src/Emogrifier/HtmlProcessor/HtmlPruner.php

@@ -2,10 +2,13 @@
 
 namespace Pelago\Emogrifier\HtmlProcessor;
 
+use Pelago\Emogrifier\Utilities\ArrayIntersector;
+
 /**
  * This class can remove things from HTML.
  *
  * @author Oliver Klee <[email protected]>
+ * @author Jake Hotson <[email protected]>
  */
 class HtmlPruner extends AbstractHtmlProcessor
 {
@@ -41,4 +44,69 @@ public function removeElementsWithDisplayNone()
 
         return $this;
     }
+
+    /**
+     * Removes classes that are no longer required (e.g. because there are no longer any CSS rules that reference them)
+     * from `class` attributes.
+     *
+     * Note that this does not inspect the CSS, but expects to be provided with a list of classes that are still in use.
+     *
+     * This method also has the (presumably beneficial) side-effect of minifying (removing superfluous whitespace from)
+     * `class` attributes.
+     *
+     * @param string[] $classesToKeep names of classes that should not be removed
+     *
+     * @return self fluent interface
+     */
+    public function removeRedundantClasses(array $classesToKeep = [])
+    {
+        $elementsWithClassAttribute = $this->xPath->query('//*[@class]');
+
+        if ($classesToKeep !== []) {
+            $this->removeClassesFromElements($elementsWithClassAttribute, $classesToKeep);
+        } else {
+            // Avoid unnecessary processing if there are no classes to keep.
+            $this->removeClassAttributeFromElements($elementsWithClassAttribute);
+        }
+
+        return $this;
+    }
+
+    /**
+     * Removes classes from the `class` attribute of each element in `$elements`, except any in `$classesToKeep`,
+     * removing the `class` attribute itself if the resultant list is empty.
+     *
+     * @param \DOMNodeList $elements
+     * @param string[] $classesToKeep
+     *
+     * @return void
+     */
+    private function removeClassesFromElements(\DOMNodeList $elements, array $classesToKeep)
+    {
+        $classesToKeepIntersector = new ArrayIntersector($classesToKeep);
+
+        foreach ($elements as $element) {
+            $elementClasses = \preg_split('/\\s++/', \trim($element->getAttribute('class')));
+            $elementClassesToKeep = $classesToKeepIntersector->intersectWith($elementClasses);
+            if ($elementClassesToKeep !== []) {
+                $element->setAttribute('class', \implode(' ', $elementClassesToKeep));
+            } else {
+                $element->removeAttribute('class');
+            }
+        }
+    }
+
+    /**
+     * Removes the `class` attribute from each element in `$elements`.
+     *
+     * @param \DOMNodeList $elements
+     *
+     * @return void
+     */
+    private function removeClassAttributeFromElements(\DOMNodeList $elements)
+    {
+        foreach ($elements as $element) {
+            $element->removeAttribute('class');
+        }
+    }
 }

src/Emogrifier/Utilities/ArrayIntersector.php

@@ -41,7 +41,7 @@ public function __construct(array $array)
      *
      * @param (int|string)[] $array
      *
-     * @return (int|string)[] Returns an array containing all of the values in `$array` whose values exist the array
+     * @return (int|string)[] Returns an array containing all of the values in `$array` whose values exist in the array
      *         with which this object was constructed.  Note that keys are preserved, order is maintained, but
      *         duplicates are removed.
      */