Support aiida-quantumespresso plugin to 4.5.0

[unkcpz]

No description provided.

[AndresOrtegaGuerrero]

For the #577 , we should wait until aiidateam/aiida-quantumespresso#1007 is merged. Also is covenient for #512

[danielhollas]

@unkcpz @superstar54 FYI this update is needed for aiida-core~=2.5 since earlier version of aiida-quantumespresso depend on pydantic v1.

This is a bit unfortunate, since you're going to have a hard breaking change in the next release --- older releases will only work on older images, while the new v2.5 image will only support the latest release.

[unkcpz]

I think pinning to the minor version will do the trick, if the aiida-core in the docker stack < 2.5, the aiida-quantumespresso==4.4 will be installed, otherwise with install aiida-quantumespresso==4.5 with pydantic v2.

[unkcpz]

Hi @superstar54, can you check why the unit test failed? I don't understand why the max_iteration is in the reference test data of data-regression output.

[superstar54]

@unkcpz this is introduced in this commit from latest aiida-quantumespresso. I think you can update the reference data to include it.

[danielhollas]

I think pinning to the minor version will do the trick, if the aiida-core in the docker stack < 2.5, the aiida-quantumespresso==4.4 will be installed, otherwise with install aiida-quantumespresso==4.5 with pydantic v2.

Sounds reasonable, I did a couple of quick tests locally to verify different combinations work, e.g.

uv pip install -e . aiida-core==2.5
uv pip install -e . aiida-core==2.4 pydantic~=1.0   # This one installes 4.4 version
uv pip install -e . aiida-core==2.4

[danielhollas]

Hi @superstar54, can you check why the unit test failed? I don't understand why the max_iteration is in the reference test data of data-regression output.

Hmm, perhaps to prevent potential breakages in the future, you might want to do

aiida-quantumespresso~=4.4,<4.6

[unkcpz]

I think you can update the reference data to include it.

No, I don't think this will solve the problem. We want our test pass for no matter which aiida-quantumespresso version is used. Ideally, we need in the qeapp to have test run on oldest and latest version of aiida-quantumespresso supported as well.

aiida-quantumespresso~=4.4,<4.6

I think this is the workaround for the moment. The max_iteration is a new thing in 4.5.0, so put <4.6 won't help.

[unkcpz]

Blocked by #638

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 75.85%. Comparing base (a6f3517) to head (17a43e2).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #601   +/-   ##
=======================================
  Coverage   75.85%   75.85%           
=======================================
  Files          60       60           
  Lines        4312     4312           
=======================================
  Hits         3271     3271           
  Misses       1041     1041           

Flag	Coverage Δ
python-3.10	75.85% <ø> (ø)
python-3.9	75.88% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[superstar54]

@danielhollas @unkcpz ready to go!

[unkcpz]

Thanks @superstar54 for the work, but I try not too much optimistic. I bump the aiida-core version in CI to 2.5, hope it passed. Please also check the comment on #644, I think the workflow data-regression test can not simply being removed but should be moved (decoupled for the workchain only).

[unkcpz]

Nice, tests still all passed after bump aiida-core version in CI matrix. Good to go! This is my PR so I can not approve. @danielhollas can you give it a final check that we didn't miss anything else.