maksaska commented on code in PR #13226:
URL: https://github.com/apache/ignite/pull/13226#discussion_r3959830754
##########
modules/ducktests/tests/ignitetest/utils/ignite_test.py:
##########
@@ -32,7 +32,7 @@
class IgniteTestContext(TestContext):
def __init__(self, test_context):
- super().__init__()
+ super().__init__(**test_context.__dict__)
self.__dict__.update(**test_context.__dict__)
Review Comment:
Thanks - checked both halves against the ducktape 0.15.0 source we're pinning
1) TypeError / filtering the kwargs
TestContext.__init__ is `def __init__(self, **kwargs)` and reads every
field via
`kwargs.get(...)`, so unknown keys are silently ignored and a TypeError
isn't
reachable — a filter would be dead code. The splat is also ducktape's own
idiom:
TestContext.copy() does `TestContext(**self.__dict__)` followed by
`ctx_copy.__dict__.update(**kwargs)`.
For context on why the line changed at all: 0.15 turned
`kwargs.get("session_context")` into `kwargs["session_context"]` and then
dereferences it (`self.session_context.enable_jvm_logs` for
ServiceRegistry).
So the previous `super().__init__()` now raises KeyError:
'session_context', not
TypeError — that's what our checks caught on the 0.15 bump
(checks/utils/check_cluster.py drives @cluster ->
ParametrizableClusterMetadata.apply
-> IgniteTestContext.resolve). Reverting to option (a) re-breaks it, so
I've left
the splat as is.
2) The subsequent self.__dict__.update(...) — agreed, fixed.
--
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]
