From cfd5d9e209d52b1d6ce18029da21e27aa2deb227 Mon Sep 17 00:00:00 2001
From: Andy Miller <rhuk@mac.com>
Date: Mon, 5 Oct 2020 18:44:59 -0600
Subject: [PATCH] improved compatibility

---
 system/src/Grav/Common/Helpers/Excerpts.php | 53 +++++++++++++++++----
 1 file changed, 43 insertions(+), 10 deletions(-)

diff --git a/system/src/Grav/Common/Helpers/Excerpts.php b/system/src/Grav/Common/Helpers/Excerpts.php
index f07c3aa5a..95e2b143a 100644
--- a/system/src/Grav/Common/Helpers/Excerpts.php
+++ b/system/src/Grav/Common/Helpers/Excerpts.php
@@ -3,7 +3,7 @@
 /**
  * @package    Grav\Common\Helpers
  *
- * @copyright  Copyright (C) 2015 - 2019 Trilby Media, LLC. All rights reserved.
+ * @copyright  Copyright (C) 2015 - 2020 Trilby Media, LLC. All rights reserved.
  * @license    MIT License; see LICENSE file for details.
  */
 
@@ -32,7 +32,7 @@ class Excerpts
         $excerpt = static::processLinkExcerpt($excerpt, $page, 'image');
 
         $excerpt['element']['attributes']['src'] = $excerpt['element']['attributes']['href'];
-        unset ($excerpt['element']['attributes']['href']);
+        unset($excerpt['element']['attributes']['href']);
 
         $excerpt = static::processImageExcerpt($excerpt, $page);
 
@@ -43,6 +43,26 @@ class Excerpts
         return $html;
     }
 
+    /**
+     * Process Grav page link URL from HTML tag
+     *
+     * @param string $html              HTML tag e.g. `<a href="../foo">Page Link</a>`
+     * @param PageInterface|null $page  Page, defaults to the current page object
+     * @return string                   Returns final HTML string
+     */
+    public static function processLinkHtml($html, PageInterface $page = null)
+    {
+        $excerpt = static::getExcerptFromHtml($html, 'a');
+
+        $original_href = $excerpt['element']['attributes']['href'];
+        $excerpt = static::processLinkExcerpt($excerpt, $page, 'link');
+        $excerpt['element']['attributes']['data-href'] = $original_href;
+
+        $html = static::getHtmlFromExcerpt($excerpt);
+
+        return $html;
+    }
+
     /**
      * Get an Excerpt array from a chunk of HTML
      *
@@ -52,22 +72,35 @@ class Excerpts
      */
     public static function getExcerptFromHtml($html, $tag)
     {
-        $doc = new \DOMDocument();
-        $doc->loadHTML($html);
-        $images = $doc->getElementsByTagName($tag);
-        $excerpt = null;
+        $doc = new \DOMDocument('1.0', 'UTF-8');
+        $internalErrors = libxml_use_internal_errors(true);
+        $doc->loadHTML(mb_convert_encoding($html, 'HTML-ENTITIES', 'UTF-8'));
+        libxml_use_internal_errors($internalErrors);
 
-        foreach ($images as $image) {
+        $elements = $doc->getElementsByTagName($tag);
+        $excerpt = null;
+        $inner = [];
+
+        /** @var \DOMElement $element */
+        foreach ($elements as $element) {
             $attributes = [];
-            foreach ($image->attributes as $name => $value) {
+            foreach ($element->attributes as $name => $value) {
                 $attributes[$name] = $value->value;
             }
             $excerpt = [
                 'element' => [
-                    'name'       => $image->tagName,
+                    'name'       => $element->tagName,
                     'attributes' => $attributes
                 ]
             ];
+
+            foreach ($element->childNodes as $node) {
+                $inner[] = $doc->saveHTML($node);
+            }
+
+            $excerpt = array_merge_recursive($excerpt, ['element' => ['text' => implode('', $inner)]]);
+
+
         }
 
         return $excerpt;
@@ -95,7 +128,7 @@ class Excerpts
 
         if (isset($element['text'])) {
             $html .= '>';
-            $html .= $element['text'];
+            $html .= is_array($element['text']) ? static::getHtmlFromExcerpt(['element' => $element['text']]) : $element['text'];
             $html .= '</'.$element['name'].'>';
         } else {
             $html .= ' />';