dhkim1920 commented on code in PR #73783:
URL: https://github.com/apache/airflow/pull/73783#discussion_r4538246663
##########
airflow-core/tests/unit/api_fastapi/execution_api/versions/head/test_task_instances.py:
##########
@@ -742,6 +744,46 @@ def
test_startup_xcom_cleanup_keys_belong_to_the_starting_try(
assert response.status_code == 200
assert response.json()["xcom_keys_to_clear"] == ["own"]
+ @pytest.mark.parametrize(
+ "conf_value",
+ [
+ pytest.param(-(2**63), id="min-int64"),
+ pytest.param(2**64 - 1, id="max-uint64"),
+ pytest.param(
+ -(2**63) - 1,
+ id="below-int64",
+ marks=pytest.mark.backend("postgres", "sqlite"),
Review Comment:
MySQL returns these out-of-range JSON integers as floats after persistence,
so the original Python integer type is not preserved, and precision can also be
lost depending on the value. This happens before MessagePack serialization, so
those two cases are restricted to PostgreSQL and SQLite. The native-range
boundary values and numeric string case still run on MySQL. I think MySQL’s
persistence behavior should be addressed separately from this transport fix.
--
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]
