Re: "Missing" entries in timezone tables

2025-04-08 Thread Michael Stone

On Tue, Apr 08, 2025 at 06:42:43AM +0100, Alastair McKinstry wrote:
But ./gen_tzinfo.py in python-tz adds extra timezones it believes 
should be present, including some backwards-compatible entries such as 
"US/Pacific". Adding these timezones is possible, but I am loath to 
diverge from tzdata..


Any opinions or recommendations?


I'd just depend on tzdata-legacy. I don't really understand why it was 
decided to put some names in a separate package, but it's a small one 
and if you need compatibility, that's what you need to do.




Re: "Missing" entries in timezone tables

2025-04-08 Thread Ananthu C V
Hi,

I am not entirely sure, but I think you can probably find those missing 
timezones in tzdata-legacy package.

Best,
Ananthu

"Missing" entries in timezone tables

2025-04-07 Thread Alastair McKinstry

Hi,

I have an issue with some missing entries in the zoneinfo files in 
tzdata that I need to resolve.


I maintain two packages, python-pytz and python-xarray.

python-tz provides a python interface to timezone data;in Debian its 
patched to read from zoneinfo rather tha ship its own copy of the data.


python-xarray has in 2025.03.1 now got a unit test that fails because 
"US/Pacific" is not present.


Now in python-tz 2025.2  it has changed its build procedure 
dramatically; the top-level Makefile pulls down zoneinfo via rsync (ugh) 
in build and edits it. In -1 I bypass this as we still apply our patch 
to use the official zoneinfo.


But ./gen_tzinfo.py in python-tz adds extra timezones it believes should 
be present, including some backwards-compatible entries such as 
"US/Pacific". Adding these timezones is possible, but I am loath to 
diverge from tzdata..


Any opinions or recommendations?

Regards

Alastair

--
Alastair McKinstry,
GPG: 82383CE9165B347C787081A2CBE6BB4E5D9AD3A5
e: [email protected], im: @alastair:mckinstry.ie
https://mastodon.ie/@amckinstry