Conversation
|
I note that, in order to get the tests passing, one need sot/proseco/pull/412, and considering this is a numpy 2 PR, I actually did my checks on the branch for #225 which is chained to this one, and rebased the proseco PR 412 on proseco PR 411 as well. Doing all that, tests pass on ska3-flight and hope. |
|
@jeanconn - this needs to be tested in ska3-flight as well. |
|
I don't generally test in ska3-flight. This is not really a numpy 2 PR at this point. |
|
#225 is not required to test this PR. |
It is not, but combinatorics is a thing. I am testing in ska3-flight and hope, and in all numpy 2 PRs. If they are chained, it makes life easier. |
|
Right, it just means you didn't test this PR in isolation and I thought the #225 code was just about warnings. |
I'm not sure what that means, but it will be faster to just test all these PRs in both flight and hope than to discuss. |
|
I do all of my ska PR testing in latest. And my latest is pre numpy 2.0 . I only documented testing above in that environment. |
|
My overall point was that since we decided to have all PRs be back compatible, my take-away was that testing in current pre-hope latest was really sufficient for development at this point and release testing will cover hope testing. But I'm typing this while I'm trying to get to ska3-flight on Linux to sure, run the tests again if ska3-flight plus the proseco PR is favored. |
|
I don't understand. I approved the PR. Do you need me to approve it again? |
|
Tom said I needed to test on ska3-flight. |
Description
Two small updates:
Updates test reference data in the test_bad_star_set test for latent bug fix found in numpy 2 testing . There should always have been the candidate criteria critical warning in the warns for that test.
Cast the star catalog index to an int in the check_guide_geometry. This is only used in the warning printing.
Requires
Interface impacts
Testing
Unit tests
Independent check of unit tests by [REVIEWER NAME]
Functional tests
No functional testing.