ProcessNode.is_valid_cache: catch AttributeError for unloadable identifier

[sphuber]

Fixes #5220

The is_valid_cache property is supposed to only raise a ValueError
when the function or class represented by the process_type identifier
cannot be loaded. However, for a normal Python identifier of an existing
module but with a function or class that does not exist in that module
an AttributeError will be thrown which was not caught and converted.

[sphuber]

@ltalirz I think this should fix your issue

[ltalirz]

Thanks a lot, I'll check it later today

[codecov]

Codecov Report

Merging #5222 (555a641) into develop (8acdc24) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@             Coverage Diff             @@
##           develop    #5222      +/-   ##
===========================================
+ Coverage    81.21%   81.21%   +0.01%     
===========================================
  Files          532      532              
  Lines        37357    37357              
===========================================
+ Hits         30336    30337       +1     
+ Misses        7021     7020       -1     

Flag	Coverage Δ
django	76.07% <100.00%> (+0.01%)	⬆️
sqlalchemy	75.17% <100.00%> (+0.01%)	⬆️

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

Impacted Files	Coverage Δ
aiida/orm/nodes/process/process.py	86.74% <100.00%> (+1.03%)	⬆️
aiida/transports/util.py	65.63% <0.00%> (-6.25%)	⬇️
aiida/transports/plugins/local.py	81.66% <0.00%> (+0.26%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8acdc24...555a641. Read the comment docs.

[ltalirz]

thanks a lot @sphuber - there's a typo but up to you whether you want to fix it.

just so I understand: the fix of the issue is implicit here, since ValueErrors will be dealt with and so the failure here is no longer catastrophic, is that right?

Shall we open a separate issue for getting a more specific process type for calcfunctions that are not defined at the module level?

[sphuber]

thanks a lot @sphuber - there's a typo but up to you whether you want to fix it.

Thanks @ltalirz away from my computer now, will take of this at the end of the weekend.

just so I understand: the fix of the issue is implicit here, since ValueErrors will be dealt with and so the failure here is no longer catastrophic, is that right?

Exactly, downstream is correctly catching ValueError but nothing else.

Shall we open a separate issue for getting a more specific process type for calcfunctions that are not defined at the module level?

Yes, thatis still an unsolved issue so good to open one.