From 74a75f7d430eb7a69cd377194e52012db34d39b4 Mon Sep 17 00:00:00 2001
From: Jeremy Benoist <jeremy.benoist@gmail.com>
Date: Fri, 12 May 2017 07:53:21 +0200
Subject: Use graby ContentExtractor to clean html

It might be better to re-use some graby functionalities to clean html instead of building a new system.
---
 src/Wallabag/ApiBundle/Controller/EntryRestController.php |  1 -
 src/Wallabag/CoreBundle/Helper/ContentProxy.php           | 10 ++++++++++
 2 files changed, 10 insertions(+), 1 deletion(-)

(limited to 'src/Wallabag')

diff --git a/src/Wallabag/ApiBundle/Controller/EntryRestController.php b/src/Wallabag/ApiBundle/Controller/EntryRestController.php
index e6bbe552..0930c109 100644
--- a/src/Wallabag/ApiBundle/Controller/EntryRestController.php
+++ b/src/Wallabag/ApiBundle/Controller/EntryRestController.php
@@ -336,7 +336,6 @@ class EntryRestController extends WallabagRestController
             $entry->setUrl($url);
         }
 
-
         if (!empty($tags)) {
             $this->get('wallabag_core.tags_assigner')->assignTagsToEntry($entry, $tags);
         }
diff --git a/src/Wallabag/CoreBundle/Helper/ContentProxy.php b/src/Wallabag/CoreBundle/Helper/ContentProxy.php
index e06ad3d6..a1df16d8 100644
--- a/src/Wallabag/CoreBundle/Helper/ContentProxy.php
+++ b/src/Wallabag/CoreBundle/Helper/ContentProxy.php
@@ -47,6 +47,16 @@ class ContentProxy
     {
         // ensure content is a bit cleaned up
         if (!empty($content['html'])) {
+            $extractor = $this->graby->getExtractor();
+            $contentExtracted = $extractor->process($content['html'], $url);
+
+            if ($contentExtracted) {
+                $contentBlock = $extractor->getContent();
+                $contentBlock->normalize();
+
+                $content['html'] = trim($contentBlock->innerHTML);
+            }
+
             $content['html'] = htmLawed($content['html'], [
                 'safe' => 1,
                 // which means: do not remove iframe elements
-- 
cgit v1.2.3