From 4aa25bf3d853d43d4e618e2197fa573851951b8c Mon Sep 17 00:00:00 2001
From: pictuga <contact@pictuga.com>
Date: Fri, 24 Mar 2017 21:50:46 -1000
Subject: [PATCH] readabilite: clean_html before scoring

Surprisingly efficient
---
 morss/readabilite.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/morss/readabilite.py b/morss/readabilite.py
index e05f68b..fcc13e9 100644
--- a/morss/readabilite.py
+++ b/morss/readabilite.py
@@ -208,13 +208,13 @@ def get_best_node(grades):
 
 def get_article(data, url=None, encoding=None):
     html = parse(data, encoding)
+
+    clean_html(html)
     br2p(html)
 
     scores = score_all(html)
     best = get_best_node(scores)
 
-    clean_html(best)
-
     if url:
         best.make_links_absolute(url)