Hi,
Hiya,
On 24.04.23 14:20, Michal Vasko wrote:
Hi,
I would like to have a specific date-and-time canonical value
clarified because I am not sure what exactly is expected. If a system
is using NZST (+12:00, New Zealand Standard Time) and is printing the
timestamp 2023-02-23T10:00:00+12:00, should the output be
1. 2023-02-23T10:00:00+12:00 because the system is currently in NZST or
2. 2023-02-23T11:00:00+13:00 because at the time of the timestamp the
system was in NZDT (+13:00)?
It is not clear to me from the ietf-yang-types:date-and-time
description and I have observed that localtime(3) adjusts the timezone
according to the timestamp meaning if I followed the same logic, I
would use 2.
RFC 3339 governs, and it's a simple rule: if the local offset at the
time is +12, use +12. If the local offset at the time is +13, use +13.
At the time, according to the normal time zone rules, it should have
been +13. But what matters more is what the device thought the offset
was *at the time*. So if it wrongly thought the offset was +12, then
that is what should be in the timestamp.
% TZ=Pacific/Auckland date -j 022311002023
Thu Feb 23 11:00:00 *NZDT* 2023.
% TZ=Pacific/Auckland date -j -Iseconds 022311002023
2023-02-23T11:00:00+*13:00*
Eliot
_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod
