From 3ecd8a6db70434a8bad3f47c6ab61d33b1b39b29 Mon Sep 17 00:00:00 2001
From: oddluck <39967334+oddluck@users.noreply.github.com>
Date: Sat, 7 Mar 2020 02:05:08 +0000
Subject: [PATCH] Jeopardy: better text reformatting

---
 Jeopardy/plugin.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/Jeopardy/plugin.py b/Jeopardy/plugin.py
index 739bfdc..60e15ad 100644
--- a/Jeopardy/plugin.py
+++ b/Jeopardy/plugin.py
@@ -276,14 +276,15 @@ class Jeopardy(callbacks.Plugin):
 
 
         def normalize(self, q):
-            q = re.sub('<[^<]+?>', '', fix_text(q, normalization='NFKC')).replace(r"\'", "'").replace(r'\"', '"')
+            q = BeautifulSoup(q)
+            q = fix_text(q.text).replace(r"\'", "'").replace(r'\"', '"')
             q = re.sub('([,;:.!?)])([a-zA-Z]|\()(?![.\'])', '\g<1> \g<2>', q)
-            q = unidecode(q)
             q = " ".join(q.split())
             return q
 
 
         def clean(self, text):
+            text = unidecode(text)
             if len(text) > 2:
                 text = re.sub('[^a-zA-Z0-9 ]+', '', text)
                 text = re.sub('^a |^an |^the |^or ', '', text).replace(' ', '')