[tz] Issues with pre-1970 information in TZDB
tgl at sss.pgh.pa.us
Thu Sep 23 17:15:01 UTC 2021
Paul Eggert <eggert at cs.ucla.edu> writes:
> On 9/22/21 12:34 PM, Tom Lane via tz wrote:
>> 1. It's not possible to separate the new backzone zones from the old.
> No, actually it is possible. I've done it, by using the patch I
> described here:
Of course I meant "tzdb, as currently constituted, provides no way
to do that without an unreasonable amount of work, as well as needing
intimate knowledge of the data set". Applying those patches would
change that state of affairs. However ...
>> 2. This approach puts it on individual tzdb distributors to decide
>> which of these two options to choose.
> Yes, that problem is inherent to any fork or option-equivalent-to-fork.
... I agree that this isn't a very desirable direction to go in,
because we really don't want different distributors shipping
different definitions of the same zone name.
I suggested nearby  that this is fundamentally caused by
misdesign of the backzone mechanism. I think a lot of the
current angst is caused by the idea that we are (depending on
build options) shipping zone definitions that are known to not
be the best available data. That's bad both intrinsically and
because it means different platforms might define the same zone
name differently. But it would be very simple to just not do that.
regards, tom lane
More information about the tz