Further restriction of possible file names, due to NAS issues

news_fetch/requirements.txt

@@ -8,4 +8,5 @@ newspaper3k
 htmldate
 markdown
 rich
-psycopg2
+psycopg2
+unidecode

news_fetch/utils_storage/helpers.py

@@ -1,7 +1,11 @@
+import unidecode
+KEEPCHARACTERS = (' ','.','_', '-')
+
 def clear_path_name(path):
-    keepcharacters = (' ','.','_', '-')
-    converted = "".join([c if (c.isalnum() or c in keepcharacters) else "_" for c in path]).rstrip()
-    return converted
+    path = unidecode.unidecode(path) # remove umlauts, accents and others
+    path = "".join([c if (c.isalnum() or c in KEEPCHARACTERS) else "_" for c in path]) # remove all non-alphanumeric characters
+    path = path.rstrip() # remove trailing spaces
+    return path
 
 def shorten_name(name, offset = 50):
     if len(name) > offset: