[tz] Issues with pre-1970 information in TZDB

[Paul Eggert]

On 9/23/21 5:53 AM, Stephen Colebourne via tz wrote:
> It is one of the possible solutions that is acceptable to me. The
> question is really "what rule determines whether an ID in the main
> data files does or does not contain pre-1970 data".

Here's one way to automate any answer that question.

1. Find out which ISO country codes even 'backzone' doesn't cover, and 
put into 'backzone' Zone entries for one location in each such country 
code. (This will require some research.)

2. Use the patch I mentioned here:

https://mm.icann.org/pipermail/tz/2021-September/030456.html

3. Write a file 'zonepercc.bp' that determines which part of 'backzone' 
you want, to support an equitable rule involving one Zone per ISO 
country code (plus all the Zones that already exist). This is as simple 
as a list of Zones you want.

4. Run 'make BACKPICK=zonepercc.bp PACKRATDATA=backzone tzdata.zi'

No doubt there are other ways to do it, but for this particular way the 
software exists already. All that's needed is some patient research and 
data selection. There's no need to fork; I'd be happy to include the 
resulting zonepercc.bp file in the tzdb code, and to discuss further 
ways to make the automation even easier.