diff roundup/backends/indexer_xapian.py @ 5964:5bf7b5debb09

Fix xapian indexer for unicode Make regular expression pattern match words as unicode.
author John Rouillard <rouilj@ieee.org>
date Wed, 30 Oct 2019 17:56:18 -0400
parents e72573996caf
children 9d209d2b34ae
line wrap: on
line diff
--- a/roundup/backends/indexer_xapian.py	Wed Oct 30 17:48:48 2019 -0400
+++ b/roundup/backends/indexer_xapian.py	Wed Oct 30 17:56:18 2019 -0400
@@ -95,7 +95,7 @@
 
         for match in re.finditer(r'\b\w{%d,%d}\b'
                                  % (self.minlength, self.maxlength),
-                                 text.upper()):
+                                 text.upper(), re.UNICODE):
             word = match.group(0)
             if self.is_stopword(word):
                 continue

Roundup Issue Tracker: http://roundup-tracker.org/