From b3fa5023c4feae49fb500f6a12b8f9d4c80f1bba Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Guilhem=20Faur=C3=A9?= <pro@gfaure.eu>
Date: Thu, 11 May 2023 10:33:35 +0200
Subject: [PATCH] fix some encoding bugs with regex replace

---
 spip2md/convert.py | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/spip2md/convert.py b/spip2md/convert.py
index 116ee4b..81a3e8a 100644
--- a/spip2md/convert.py
+++ b/spip2md/convert.py
@@ -97,10 +97,23 @@ mappings = (
         ),
         r"\1",
     ),
+    (  # Fix UTF-8 appostrophe that was interpreted as ISO 8859-1 and saved like so
+        re.compile("’"),
+        r"’",
+    ),
+    (  # Fix UTF-8 é that was interpreted as ISO 8859-1 and saved like so
+        re.compile("eÌ\u0081"),
+        r"é",
+    ),
+    (  # Fix UTF-8 é that was interpreted as ISO 8859-1 and saved like so
+        re.compile("à"),
+        r"à",
+    ),
 )
 
 
 def convert(markup):
     for spip, markdown in mappings:
         markup = spip.sub(markdown, markup)
-    return markup.encode("utf-8").decode("utf-8")
+    # return markup.encode("utf-8").decode("utf-8")
+    return markup