um7: depend on oasis3-mct branch access-esm1.5

[harshula]

No description provided.

[penguian]

The comparison ACCESS-NRI/oasis3-mct@access-esm1.5...ACCESS-NRI:oasis3-mct:access-esm1.5-new-modules shows non-trivial differences between the two branches. Are each of these differences intended, and what are their purpose?

[penguian]

Additionally, my first attempts at installation failed:

[pcl851@gadi-login-09 spack-packages]$ spack install um7%intel
==> Error: concretization failed for the following reasons:

   1. Cannot satisfy '[email protected]'
   2. Cannot satisfy '[email protected]'
        required because um7 depends on [email protected] 
          required because um7%intel requested explicitly 
[pcl851@gadi-login-09 spack-packages]$ spack install [email protected]%intel
==> Error: access-esm1.5 is not a valid git ref for oasis3-mct
[pcl851@gadi-login-09 spack-packages]$ git status -uno
On branch development
Your branch is up to date with 'origin/development'.

nothing to commit (use -u to show untracked files)

[harshula]

Hi @penguian , I've updated the access-esm1.5 branch. Can you please verify again?

[harshula]

spack install [email protected]%intel
==> Error: access-esm1.5 is not a valid git ref for oasis3-mct

# spack install [email protected] %[email protected] arch=linux-rocky8-x86_64
...
==> Installing oasis3-mct-git.access-esm1.5=0-git.71-vuzjsdgqpjwgb55nb2gwhiafaqtqtfl5
==> No binary for oasis3-mct-git.access-esm1.5=0-git.71-vuzjsdgqpjwgb55nb2gwhiafaqtqtfl5 found: installing from source
==> No patches needed for oasis3-mct
==> oasis3-mct: Executing phase: 'edit'
==> oasis3-mct: Executing phase: 'build'
==> oasis3-mct: Executing phase: 'install'
==> oasis3-mct: Successfully installed oasis3-mct-git.access-esm1.5=0-git.71-vuzjsdgqpjwgb55nb2gwhiafaqtqtfl5
  Stage: 2.55s.  Edit: 0.00s.  Build: 2m 4.96s.  Install: 0.03s.  Post-install: 0.19s.  Total: 2m 7.89s
[+] /opt/release/linux-rocky8-x86_64/intel-2021.2.0/oasis3-mct-git.access-esm1.5=0-git.71-vuzjsdgqpjwgb55nb2gwhiafaqtqtfl5

[aidanheerdegen]

The Ruff checks should be addressed. If we're not using pprint then we shouldn't import it. The f-string errors are just pointing out that there are no variable interpolations in those strings, so they can just be simple strings without the f.

[harshula]

Hi @aidanheerdegen , Will address linter issues when adding support for oneapi compilers.

[aidanheerdegen]

Ok to leave linter changes for now.