Added normalizeEntities flag.

1 files changed, 6 insertions, 0 deletions

diff --git a/src/HTMLParser.php b/src/HTMLParser.php
index 3226904..105e50c 100644
--- a/src/HTMLParser.php
+++ b/src/HTMLParser.php
@@ -100,6 +100,7 @@ class HTMLParser
             'removeReadabilityTags' => true,
             'fixRelativeURLs' => false,
             'substituteEntities' => true,
+            'normalizeEntities' => false,
             'originalURL' => 'http://fakehost',
         ];
 
@@ -208,6 +209,11 @@ class HTMLParser
             $this->dom->substituteEntities = false;
         }
 
+        if ($this->getConfig()->getOption('normalizeEntities')) {
+            // Replace UTF-8 characters with the HTML Entity equivalent. Useful to fix html with mixed content
+            $html = mb_convert_encoding($html, 'HTML-ENTITIES', 'UTF-8');
+        }
+
         // Prepend the XML tag to avoid having issues with special characters. Should be harmless.
         $this->dom->loadHTML('<?xml encoding="UTF-8">' . $html);
         $this->dom->encoding = 'UTF-8';