With 2009n, zdump shows that DST was introduced for 2009 and does not continue beyond 2009, and shows UTC+6 for non-DST (21600 seconds), and UTC+7 for DST (25200 seconds):
$ zdump -v -c 2007,2015 Asia/Dhaka
Asia/Dhaka Fri Jun 19 16:59:59 2009 UTC = Fri Jun 19 22:59:59 2009 BDT isdst=0 gmtoff=21600
Asia/Dhaka Fri Jun 19 17:00:00 2009 UTC = Sat Jun 20 00:00:00 2009 BDST isdst=1 gmtoff=25200
Asia/Dhaka Thu Dec 31 16:59:59 2009 UTC = Thu Dec 31 23:59:59 2009 BDST isdst=1 gmtoff=25200
Asia/Dhaka Thu Dec 31 17:00:00 2009 UTC = Thu Dec 31 23:00:00 2009 BDT isdst=0 gmtoff=21600
With 2009o, the bug manifests so that DST is not considered at all:
$ zdump -v -c 2007,2015 Asia/Dhaka
Asia/Dhaka Fri Jun 19 16:59:59 2009 UTC = Fri Jun 19 22:59:59 2009 BDT isdst=0 gmtoff=21600
Asia/Dhaka Fri Jun 19 17:00:00 2009 UTC = Fri Jun 19 23:00:00 2009 BDST isdst=0 gmtoff=21600
Meh, sorry for the mess. Should have checked that earlier.
With 2009n, zdump shows that DST was introduced for 2009 and does not continue beyond 2009, and shows UTC+6 for non-DST (21600 seconds), and UTC+7 for DST (25200 seconds):
$ zdump -v -c 2007,2015 Asia/Dhaka
Asia/Dhaka Fri Jun 19 16:59:59 2009 UTC = Fri Jun 19 22:59:59 2009 BDT isdst=0 gmtoff=21600
Asia/Dhaka Fri Jun 19 17:00:00 2009 UTC = Sat Jun 20 00:00:00 2009 BDST isdst=1 gmtoff=25200
Asia/Dhaka Thu Dec 31 16:59:59 2009 UTC = Thu Dec 31 23:59:59 2009 BDST isdst=1 gmtoff=25200
Asia/Dhaka Thu Dec 31 17:00:00 2009 UTC = Thu Dec 31 23:00:00 2009 BDT isdst=0 gmtoff=21600
With 2009o, the bug manifests so that DST is not considered at all:
$ zdump -v -c 2007,2015 Asia/Dhaka
Asia/Dhaka Fri Jun 19 16:59:59 2009 UTC = Fri Jun 19 22:59:59 2009 BDT isdst=0 gmtoff=21600
Asia/Dhaka Fri Jun 19 17:00:00 2009 UTC = Fri Jun 19 23:00:00 2009 BDST isdst=0 gmtoff=21600
Meh, sorry for the mess. Should have checked that earlier.