From e3508e365f72a12608badcae997862d43bbad0f8 Mon Sep 17 00:00:00 2001
From: Matthew Scragg <matt.scragg@dsisolutions.biz>
Date: Tue, 8 Dec 2015 17:15:11 -0600
Subject: [PATCH] allow utf8 in page names

---
 realms/lib/util.py            | 16 ++++++++++------
 realms/modules/wiki/models.py |  2 +-
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/realms/lib/util.py b/realms/lib/util.py
index 5eed026..2acb402 100644
--- a/realms/lib/util.py
+++ b/realms/lib/util.py
@@ -86,14 +86,18 @@ def to_canonical(s):
     Double space -> single dash
     Double dash -> single dash
     Remove all non alphanumeric and dash
-    Limit to first 64 chars
+    Limit to first 128 chars
     """
-    s = s.encode('ascii', 'ignore')
+    reserved_chars = "&$+,/:;=?@#"
+    unsafe_chars = "?<>[]{}|\^~%"
+
+    s = s.encode('utf8')
     s = str(s)
-    s = re.sub(r"\s\s*", "-", s)
-    s = re.sub(r"\-\-+", "-", s)
-    s = re.sub(r"[^a-zA-Z0-9\-]", "", s)
-    s = s[:64]
+    s = s.strip()
+    s = re.sub(r"\s", "-", s)
+    s = re.sub(r"[" + re.escape(reserved_chars) + "]", "", s)
+    s = re.sub(r"[" + re.escape(unsafe_chars) + "]", "", s)
+    s = s[:128]
     return s
 
 
diff --git a/realms/modules/wiki/models.py b/realms/modules/wiki/models.py
index 27b0852..4a2a8e7 100644
--- a/realms/modules/wiki/models.py
+++ b/realms/modules/wiki/models.py
@@ -183,7 +183,7 @@ class Wiki(HookMixin):
             return cached
 
         # commit = gittle.utils.git.commit_info(self.repo[sha])
-        filename = cname_to_filename(name).encode('latin-1')
+        filename = cname_to_filename(name).encode('utf8')
         sha = sha.encode('latin-1')
 
         try: