[justinpakzad]

---

The serde timezone serializer can't handle timezones that are just a fixed offset with no name. The datetime serializer saves an empty place holder and reading the value back (e.g., via Xcom pull) fails. Take the following Dag:

from datetime import datetime
from airflow.sdk import dag, task

@dag
def serde_offset_tz():
    @task
    def push_offset():
        return datetime.fromisoformat("2024-01-01T00:00:00+02:00")
    @task
    def do_something(value):
        print(repr(value), value.utcoffset())

    do_something(push_offset())
serde_offset_tz()

On main, do_something would fail with: ValueError: ZoneInfo keys must be normalized relative paths, got:

This PR adds support for timezones with a fixed, nonzero UTC offset and no name, by encoding them as a pendulum FixedTimezone. Additional tests were added to cover the new behavior. The NoNameTZ helper now returns None from utcoffset so that test_timezone_serialize_no_name still covers the fallback path (no name or fixed offset).

Was generative AI tooling used to co-author this PR?

• Yes (Claude Sonnet 5.5)

---

• Read the Pull Request Guidelines for more information. Note: commit author/co-author name and email in commits become permanently public when merged.
• For fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
• When adding dependency, check compliance with the ASF 3rd Party License Policy.
• For significant user-facing changes create newsfragment: {pr_number}.significant.rst, in airflow-core/newsfragments. You can add this file in a follow-up commit after the PR is created so you know the PR number.

[fat-catTW]

Thanks for the PR. Look good overall!
Non-blocking: could we also assert that the fixed offset itself survives the datetime round-trip?

test_datetime currently compares only timestamp(), so a value restored as 07:00+00:00 would still pass for an original 09:00+02:00.
An assertion such as the following would make the intended guarantee explicit:

assert deserialized_obj.utcoffset() == datetime.timedelta(hours=2)

[justinpakzad]

Thanks for the PR. Look good overall! Non-blocking: could we also assert that the fixed offset itself survives the datetime round-trip?

test_datetime currently compares only timestamp(), so a value restored as 07:00+00:00 would still pass for an original 09:00+02:00. An assertion such as the following would make the intended guarantee explicit:

Yep! Good call. The else branch actually never even ran here because they all matched the first isinstance check, so the .timestamp() comparison never happened. I've updated the test to check utcoffset() for datetimes and == for everything else.

[shahar1]

Unnamed fixed-offset tzinfo (e.g. datetime.timezone(timedelta(hours=2))) now round-trips through serde as a pendulum FixedTimezone, matching how the existing UTC special case already worked, and the new parametrized case plus test_timezone_serialize_fixed_offset both fail on main without the fix. The extra utcoffset() assertion in test_datetime is a good catch too, since == on aware datetimes alone would not notice a lost offset.

---

This review was drafted by an AI-assisted tool and
confirmed by an Apache Airflow maintainer. The maintainer
approving this PR has read the findings and signed off. If
something feels off, please reply on the PR and a maintainer
will follow up.

More on how Apache Airflow handles maintainer review:
Contributing guide.

[github-actions]

Backport successfully created: v3-4-test

Note: As of Merging PRs targeted for Airflow 3.X
the committer who merges the PR is responsible for backporting the PRs that are bug fixes (generally speaking) to the maintenance branches.

In matter of doubt please ask in #release-management Slack channel.

Status	Branch	Result
✅	v3-4-test