From 71a5fd3522c3ca6bf0ff3b5f5246f0fc8c8bca5d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tom=C3=A1=C5=A1=20Ml=C3=A1dek?= <t@mldk.cz>
Date: Thu, 23 May 2019 14:36:12 +0200
Subject: [PATCH] unicode normalize in sanitize function

---
 delojza.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/delojza.py b/delojza.py
index 507dc1a..004d42b 100755
--- a/delojza.py
+++ b/delojza.py
@@ -8,6 +8,7 @@ import re
 import shutil
 import sys
 import tempfile
+import unicodedata
 from configparser import ConfigParser
 from datetime import datetime, timedelta
 from glob import glob
@@ -92,10 +93,12 @@ class DelojzaBot:
                 return True
         return False
 
+    # https://github.com/django/django/blob/master/django/utils/text.py#L393
     @staticmethod
     def sanitize(filepath):
         if filepath is None:
             return None
+        filepath = unicodedata.normalize('NFKD', filepath).encode('ascii', 'ignore').decode('ascii')
         return re.sub(r'[^\w.()\[\]{}#-]', '_', filepath)
 
     def tag_file(self, filepath, message, info=None):