Popup Calendar showing a date twice, on different weekdays
|Assignee:||Holger Just||% Done:|
Attatched a screenshot, showing october, 17 on saturday AND sunday.
-- Redmine 0.8.3.stable (MySQL)
#4 Updated by Guilherme Schneider over 10 years ago
- File outubro-2010.png added
It happend to me too, here in Brazil. See attachment (from the redmine site, version 1.0.2 I guess?)
It must be related to DST (Daylight saving time), because here in Brazil, the DST begun at October 17th at midnight (the clock goes back to October 16th at 11:00 PM.
I found out that this bug has been resolved in JSCal2-1.8, see here: http://www.dynarch.com/projects/calendar/download/1.8/
#6 Updated by Holger Just over 10 years ago
- Assignee set to Holger Just
I can not reproduce the bug on either Redmine Trunk, nor the redmine.org Redmine. even if I set the language to Português (Brazil) and the timezone to (GMT -03:00) Brasilia. Everything looks right.
Could you please state exactly under which circumstances and which user settings you observe the bug?
#8 Updated by Holger Just over 10 years ago
- Affected version (unused) changed from 0.8.3 to devel
- Affected version deleted (
#9 Updated by Holger Just over 10 years ago
I have a patch at http://github.com/meineerde/redmine/commit/aaede7f75501c6314328fbf50f776f162346e81c. This one fixes the issue for me. Please check it.
The essence of the patch is, to use 12:00:00 as a reference time for dates instead of 00:00:00. Thus, there can't be any disambiguity on the date (at least, if there is no DST rule somewhere in the world where they change time at noon).
@Eric: Once this is confirmed, I'm going to send you a pull request.
#13 Updated by Fernando Silveira over 8 years ago
I'm using Redmine 2.0.3.stable and 1.4.4.stable and both have the same bug. After applying the patch from Holger (https://github.com/meineerde/redmine/commit/aaede7f75501c6314328fbf50f776f162346e81c) the bug is gone. Can someone PLEASE commit his pull request?
Holger, could you please ask someone you know to apply your commit? Maybe you could assign this issue to the committer, I don't know.