Filter for milestone doesn't work with umlauts

[Rochi]

The filter for Milestones doesn't work with german umlauts.

milestone=SäHO

doesn't work

[bill5753@…]

I had confronted with the same problem, milestone=研发中心, when the milestone's name is Chinese, JS Gantt will not work.

[bill5753@…]

I found a solution for that, as long as you can use the TracQuery syntax as a filter, you can sort the specific tickets by keywords which written in ascii character in advance, so that you can have all things done

[bill5753@…]

I haved solved that problem, the patch will coming soon.

[Chris Nelson]

I can reproduce this but a patch would be welcome.

[Chris Nelson]

This problem appears to arise from

        # Construct the querystring. 
        query_string = '&'.join(['%s=%s' % 
                                 (f, str(v)) for (f, v) in 
                                 query_args.iteritems()]) 

which forces values to (ASCII) strings.

[falkb]

I had this problem with German umlauts too, and this one has helped:

tracjsganttplugin/0.11/tracjsgantt/tracjsgantt.py

@@ -3 +3 @@
 from datetime import timedelta, datetime
 from operator import itemgetter, attrgetter
 
+from trac.util.text import to_unicode
 from trac.util.html import Markup
 from trac.util.text import javascript_quote
 from trac.wiki.macros import WikiMacroBase
@@ -299 +300 @@
 
         # Construct the querystring. 
         query_string = '&'.join(['%s=%s' % 
-                                 (f, str(v)) for (f, v) in 
+                                 (f, unicode(v)) for (f, v) in 
                                  query_args.iteritems()]) 
 
         # Get the Query Object. 

[Chris Nelson]

(In [11740]) Allow unicode field values (e.g,. for accents in milestones). Refs #9999.

Thanks to falkb for the patch.