search

spacetelescope / roman_datamodels

Datamodel support for the roman calibration pipeline
https://roman-datamodels.readthedocs.io
Other
7 stars 21 forks source link

replace usages of ``copy_arrays`` with ``memmap`` #360

Closed zacharyburnett closed 2 months ago

zacharyburnett commented 3 months ago

follow-on to https://github.com/asdf-format/asdf/pull/1797

Checklist

codecov[bot] commented 3 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 97.20%. Comparing base (087a60d) to head (7608ae5). Report is 29 commits behind head on main.

Additional details and impacted files ```diff @@ Coverage Diff @@ ## main #360 +/- ## ========================================== - Coverage 97.56% 97.20% -0.36% ========================================== Files 30 37 +7 Lines 2788 3359 +571 ========================================== + Hits 2720 3265 +545 - Misses 68 94 +26 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

braingram commented 3 months ago

Bumping the asdf version to at least 3.1.0 should fix the oldest deps: https://github.com/spacetelescope/roman_datamodels/blob/1b298e10221c293fe82fdc7d75603da9e816e89f/pyproject.toml#L16

PaulHuwe commented 2 months ago

Can you please add a RCAL regression test URL as well?

zacharyburnett commented 2 months ago

Can you please add a RCAL regression test URL as well?

RCAL regtest running at https://github.com/spacetelescope/RegressionTests/actions/runs/10042547798 with this branch installed

zacharyburnett commented 2 months ago

a few failures in RCAL regtests: https://github.com/spacetelescope/RegressionTests/actions/runs/10042547798/attempts/2#summary-27759909972

braingram commented 2 months ago

a few failures in RCAL regtests: https://github.com/spacetelescope/RegressionTests/actions/runs/10042547798/attempts/2#summary-27759909972

The failures are unrelated (they're expected and caused by https://github.com/spacetelescope/romancal/pull/1241).

Would you re-run them? I okified the results from a different run but I think that wasn't until after your tests had run.

zacharyburnett commented 2 months ago

sure, rerunning here: https://github.com/spacetelescope/RegressionTests/actions/runs/10042547798

braingram commented 2 months ago

The oldest deps failure should be solvable with a bump to the gwcs version: https://github.com/spacetelescope/roman_datamodels/actions/runs/10045391464/job/27762571944?pr=360#step:10:292

braingram commented 2 months ago

Sorry for the back and forth but this will likely need another run. It turns out that some of the roman tests continue to fail after the first "okify": https://github.com/spacetelescope/romancal/issues/1325 I have a run on main to test if the current "okified" results pass. I'll post and update here once the run on main passes.

braingram commented 2 months ago

I re-ran the regtests for this PR (now that the romancal main ones are passing): https://github.com/spacetelescope/RegressionTests/actions/runs/10042547798

The run with this PR shows all passing tests.