Time zone data for Dhaka is missing an end date

To whom it may concern, I am writing to report an issue I have discovered in the recent change to the Asia/Dhaka time zone data, which recently implemented daylight saving time in June. The recent update to the time zone data (tzdata2009j) includes a start date for daylight saving time, but no end date. Intuitively, daylight saving time should have a start and an end date within the same year, otherwise what is implemented is an indefinite GMT offset change. Because of this programs that detect DST rules may see the absence of an end to the DST period as an error. This may cause programs which obtain DST rules using Olson time zone data to fail to detect daylight saving time rules for Asia/Dhaka. For completeness, I propose a tentative end date be implemented now, and an updated date be implemented once an official date is released, or the government releases an additional tentative date in September. According to timeanddate.com: "The press secretary did not confirm the length of DST but stated that the government would decide the DST end date and would announce it later. On another note, Bangladesh's minister for power said earlier in 2009 that the new timing might continue until the end of September. Timeanddate.com will provide an update on the DST end date when it becomes available. " ( http://www.timeanddate.com/news/time/bangladesh-daylight-saving-2009.html ): Timeanddate.com picked September 25th, which seems like a likely enough date. I have update the tzdata on my system with the following rules to implement this change. # Rule NAME FROM TO TYPE IN ON AT SAVE LETTER/S Rule Dhaka09 2009 max - Jun 19 23:00 1:00 D Rule Dhaka09 2009 max - Sep 26 00:00 0:00 S # Zone NAME GMTOFF RULES FORMAT [UNTIL] Zone Asia/Dhaka 6:01:40 - LMT 1890 5:53:20 - HMT 1941 Oct # Howrah Mean Time? 6:30 - BURT 1942 May 15 # Burma Time 5:30 - IST 1942 Sep 6:30 - BURT 1951 Sep 30 6:00 - DACT 1971 Mar 26 # Dacca Time 6:00 Dhaka09 B%sT # Bangladesh Time Could you please evaluate this change for inclusion in the Olson time zone data? Please let me know if you have any questions or concerns. Thanks, Nik
participants (1)
-
nikolaus_brauer@us.ibm.com