On 10/30/2015 01:42 PM, Random832 wrote:
My message disappeared. I've determined this was because I changed my primary email address and wasn't recognized as subscribed to the list, but it would have been nice to have received a notification of this as I have for other lists.
On Thu, Oct 29, 2015, at 16:14, Paul Eggert wrote:
Sorry, I must be missing something, but I am not following the question. For that absolute transition, there's no difference between 02:00 standard time, 03:00 local time, and 08:00 UTC; they're all the same thing, and the "s" in "2:00s" is purely for convenience in the source data. The binary data does include information about the difference for the purpose of determining how the transition should be treated if the timezone is transposed to a different offset due to being used as the basis for interpreting POSIX rules.
This data is in the sections which tzh_ttisgmtcnt/tzh_ttisstdcnt determines the size of.
Right. Is this info actually used in practice? If not, then I probably don't care. Anyhow, back to my original question, I still don't understand why zic doesn't flag the Feb 9, 1942 transition as being observed at standard time. Zone America/Atikokan -6:06:28 - LMT 1895 -6:00 Canada C%sT 1940 Sep 29 -6:00 1:00 CDT 1942 Feb 9 2:00s -6:00 Canada C%sT 1945 Sep 30 2:00 -5:00 - EST As part of my tzdist server, I've written code to translate iCalendar VTIMEZONE data back into tzfile format, and the only place that my files differ from files compiled with zic are those indicators in which an UNTIL time has a non-wall time suffix. All my transition times and other indicators match. For the zone above I have the ttisstd bit set in my data for the Feb 9, 1942 transition. zic (2015g) does not set this bit. -- Kenneth Murchison Principal Systems Software Engineer Carnegie Mellon University