Idsse 1054

[paulhamer-noaa]

Linear Issue

IDSSE-1054

Changes

• Some updates to code for linter issues
• Increase coverage for rabbitmq_utils (only to 88% still require additional testing)

Explanation

Created pull request to check-point work so far. Consider splitting rabbitmq_utils, i.e. Rpc, Consumer, Publisher classes as separate modules to reduce module sizes.

[Geary-Layne]

Do I understand correctly, with these changes we are close to 90% test coverage but still under? If so, I want to make sure we create an additional task to come back and increase the coverage.
A number of these tests directly call the private _publish function. If you are able to make changes to call one of the public publish functions (that will internally call the private function) that would be better. If not, we need to make another task to revisit these test.

[Geary-Layne]

I would rather we not call 'private' functions directly in tests. This practice often makes it more work if/when we need to refactor our code. It is better to only call 'public' functions, though I know it can sometimes be difficult to work out the correct args to the public functions that test the underlying core logic.

[paulhamer-noaa]

I did make tests to call some of the _private functions but, yes, an additional tak to bring it up to 90% would be good.

…

On Thu, Jan 2, 2025 at 7:21 AM Geary Layne ***@***.***> wrote: ***@***.**** approved this pull request. Do I understand correctly, with these changes we are close to 90% test coverage but still under? If so, I want to make sure we create an additional task to come back and increase the coverage. A number of these tests directly call the private _publish function. If you are able to make changes to call one of the public publish functions (that will internally call the private function) that would be better. If not, we need to make another task to revisit these test. ------------------------------ On python/idsse_common/test/test_rabbitmq_utils.py <#91 (comment)> : I would rather we not call 'private' functions directly in tests. This practice often makes it more work if/when we need to refactor our code. It is better to only call 'public' functions, though I know it can sometimes be difficult to work out the correct args to the public functions that test the underlying core logic. — Reply to this email directly, view it on GitHub <#91 (review)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ATO5PJIXXP2KLL63WYE46BL2IVDPBAVCNFSM6AAAAABUN6HFI6VHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZDKMRXGY2DEMBUGQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Paul Hamer <http://www.esrl.noaa.gov/fiqas/> Global Systems Laboratory (GSL) 325 Broadway R/GS5 Boulder, CO 80305-3328