From d8471fc68e363ac884591cf9ab9e334773f97cae Mon Sep 17 00:00:00 2001
From: Andres Rey <andreskrey@gmail.com>
Date: Sat, 1 Sep 2018 20:50:25 +0100
Subject: Import the isPhrasingContent function. Might want to check the
 recursive loop there if it's actually doing what it should and if there's a
 better way to optimize it

---
 src/Nodes/NodeTrait.php | 32 ++++++++++++++++++++++++++++++++
 src/Readability.php     |  4 ++++
 2 files changed, 36 insertions(+)

diff --git a/src/Nodes/NodeTrait.php b/src/Nodes/NodeTrait.php
index 13611c9..9f24abc 100644
--- a/src/Nodes/NodeTrait.php
+++ b/src/Nodes/NodeTrait.php
@@ -50,6 +50,21 @@ trait NodeTrait
         'select',
     ];
 
+    /**
+     * The commented out elements qualify as phrasing content but tend to be
+     * removed by readability when put into paragraphs, so we ignore them here.
+     *
+     * @var array
+     */
+    private $phrasing_elems = [
+        // 'CANVAS', 'IFRAME', 'SVG', 'VIDEO',
+        'abbr', 'audio', 'b', 'bdo', 'br', 'button', 'cite', 'code', 'data',
+        'datalist', 'dfn', 'em', 'embed', 'i', 'img', 'input', 'kbd', 'label',
+        'mark', 'math', 'meter', 'noscript', 'object', 'output', 'progress', 'q',
+        'ruby', 'samp', 'script', 'select', 'small', 'span', 'strong', 'sub',
+        'sup', 'textarea', 'time', 'var', 'wbr'
+    ];
+
     /**
      * initialized getter.
      *
@@ -431,4 +446,21 @@ trait NodeTrait
 
             );
     }
+
+    /**
+     * Determine if a node qualifies as phrasing content.
+     * https://developer.mozilla.org/en-US/docs/Web/Guide/HTML/Content_categories#Phrasing_content
+     *
+     * @return bool
+     */
+    public function isPhrasingContent()
+    {
+        return $this->nodeType === XML_TEXT_NODE || !in_array($this->tagName, $this->phrasing_elems) ||
+            (!is_null($this->childNodes) &&
+                ($this->tagName === 'a' || $this->tagName === 'del' || $this->tagName === 'ins') &&
+                array_reduce(iterator_to_array($this->childNodes), function ($carry, $node) {
+                    return $carry || $node->isPhrasingContent();
+                })
+            );
+    }
 }
diff --git a/src/Readability.php b/src/Readability.php
index 3f4d807..e0c9abf 100644
--- a/src/Readability.php
+++ b/src/Readability.php
@@ -831,6 +831,10 @@ class Readability
                         }
                     }
 
+                    if ($next->isPhrasingContent()) {
+                        break;
+                    }
+
                     $this->logger->debug('[PrepDocument] Replacing BR with a P node...');
 
                     // Otherwise, make this node a child of the new <p>.
-- 
cgit v1.2.3