[tz] Weird PST8PDT and EST5EDT behavior on Alpine Linux

[Guy Harris]

On Mar 6, 2024, at 7:39 AM, James Cloos <cloos at jhcloos.com> wrote:

>>>>>> "RA" == Russ Allbery <eagle at eyrie.org> writes:
> 
> RA> and it's probably time to assume everyone supports the Olson
> RA> identifiers.
> 
> Operwrt, at least in general, does not.

I.e., it doesn't provide tzdb files, to save storage space on small embedded devices?

> (Nor does aix, if anyone still cares.)

According to at least one page on IBM's support web site, "The default timezone format for AIX 6.1 and AIX 7 is Olson Time":

	https://www.ibm.com/support/pages/managing-time-zone-variable-posix

Is that no longer the case?

> I have a clock (actual hw clock) which uses ntp by way of an embedded
> mips board running a version of openwrt.  It lack the ram and storage
> to use tzdb.  So I have to use EST5EDT,M3.2.0/2:00:00,M11.1.0/2:00:00
> for the timezone.  (The explicit /2:00:00 might be avoidable, but the
> use of self documenting data can be beneficial.)

If the desire here is to allow all users of a system to specify time zones using old-style TZ strings such as EST5EDT rather than either POSIX-style or tzdb-style strings, if the system in question can be used outside of a particular hardwired polity, that'll require some way for time zone rules to be specified.

If the desire here is to allow old-style TZ strings within a given polity - said polity probably being the US - to use old-style TZ strings, that could be done by hardcoding the rules for the US and use them, and, if the US changes the rules, distribute an update to the system.

> (I must say it was a bit anoying that the library kept the original
> definition of EST5EDT et alia rathar than moving with the legislation.
> But only a bit.)

To which library are you referring here?