On 2021-06-02 17:24, John Hawkinson via tz wrote:
Brian Inglis via tz <tz@iana.org> wrote on Wed, 2 Jun 2021 at 13:10:34 EDT in <974cf8e1-c2d8-36bf-ec10-8e99ec826956@SystematicSw.ab.ca>:
@-67768040609740800 -2147481748 Jan 01 Mon 00:00:00 @67767976233532799 2147483647 Dec 31 Tue 23:59:59 That range isn't part of GNU date, I don't think. Under OS X, I get a different result. For the negative bound:
pb3:xj2 jhawk$ gdate --version date (GNU coreutils) 8.30 ... pb3:xj2 jhawk$ gdate -d @-67768040609722801 gdate: time ‘-67768040609722801’ is out of range pb3:xj2 jhawk$ gdate -d @-67768040609722800 Thu Jan 1 00:00:00 EST -2147481748 pb3:xj2 jhawk$
The value you have given, 1970-01-01 - 67 768 040 609 722 800 seconds, is the smallest value of UT for which the EST value U - 05 h is representable in struct tm. Brian Inglis has given the smallest and largest datetimes representable in a struct tm; this shifts the range by 5 hours. What I find surprising is that the days of the week as given by Brian are incorrect: the Gregorian dates -2147 481 748-01-01 +2147 485 548-01-01 are both Thursdays. Michael Deckers.