[unicode] Wide version of word breaks

scripts/geodata/i18n/word_breaks.py

@@ -49,12 +49,7 @@ ideographic_scripts = set([
 def regex_char_range(match):
     r = match.split('..')
     # Wide version
-    # return u'-'.join([('\u{}' if len(c) < 5 else '\U{}').format(c) for c in r])
-
-    if len(r[0]) < 5 and len(r[-1]) < 5:
-        return '-'.join(['\u{}'.format(c.lower()) for c in r])
-    else:
-        return ''
+    return u'-'.join([('\u{}'.format(c.lower()) if len(c) < 5 else '\U{}'.format(c.lower().rjust(8, '0'))) for c in r])
 
 
 def get_letter_range(text, *regexes):