burakcoleman opened a new pull request, #70469:
URL: https://github.com/apache/airflow/pull/70469
`DateTimeSensorAsync.__init__` computes
`start_trigger_args.trigger_kwargs["moment"]` by calling
`timezone.parse(self.target_time)` at Dag-parse time, before Jinja
templating has run. `target_time`
is a documented template field (e.g. `"{{
data_interval_end.tomorrow().replace(hour=1) }}"`), so a
templated value raises a raw `pendulum.parsing.exceptions.ParserError`
there, crashing parsing of the
entire Dag file, not just the one task.
This can't be fixed by moving the parse into `execute()`: when
`start_from_trigger=True`, the scheduler
defers the task straight from `TaskInstance.defer_task()` using the
pre-computed `self.task.start_trigger_args`
— `execute()` never runs for this path. So the value genuinely has to be
resolved in `__init__`, the same
constraint `TimeSensor` hit in #69610.
Fix: catch the parse failure and raise a clear `ValueError` explaining that
`start_from_trigger=True`
requires a static `target_time`, instead of letting the raw `ParserError`
propagate and take down the
whole Dag file. Static/ISO `target_time` values are unaffected and keep
working exactly as before.
Added tests covering both the static-time (still works) and templated-time
(now raises a clear,
task-scoped error instead of crashing Dag parsing) cases.
closes: #70284
Related: #69610
##### Was generative AI tooling used to co-author this PR?
- [X] Yes — Claude Code
Generated-by: Claude Code following [the
guidelines](https://github.com/apache/airflow/blob/main/contributing-docs/05_pull_requests.rst#gen-ai-assisted-contributions).
I identified the root cause, wrote the fix and tests, ran `ruff check`, `ruff
format --check`, `mypy`, and the full test file locally (12 passed), and
verified the fix against the original repro before submitting.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
