Add validation of filepaths for non-BIDS NWB assets

[jwodder]

Closes #1163.

[codecov]

Codecov Report

Base: 88.92% // Head: 89.16% // Increases project coverage by +0.23% 🎉

Coverage data is based on head (03854a7) compared to base (a0b0015).
Patch coverage: 98.55% of modified lines in pull request are covered.

❗ Current head 03854a7 differs from pull request most recent head 55df999. Consider uploading reports for the commit 55df999 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1173      +/-   ##
==========================================
+ Coverage   88.92%   89.16%   +0.23%     
==========================================
  Files          76       76              
  Lines        9393     9506     +113     
==========================================
+ Hits         8353     8476     +123     
+ Misses       1040     1030      -10     

Flag	Coverage Δ
unittests	89.16% <98.55%> (+0.23%)	⬆️

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

Impacted Files	Coverage Δ
dandi/files/_private.py	98.11% <80.00%> (ø)
dandi/cli/tests/test_cmd_validate.py	100.00% <100.00%> (ø)
dandi/cli/tests/test_command.py	100.00% <100.00%> (ø)
dandi/files/__init__.py	93.75% <100.00%> (+15.26%)	⬆️
dandi/files/bases.py	77.30% <100.00%> (+0.57%)	⬆️
dandi/organize.py	83.54% <100.00%> (+0.73%)	⬆️
dandi/tests/fixtures.py	98.60% <100.00%> (+0.03%)	⬆️
dandi/tests/test_files.py	100.00% <100.00%> (ø)
dandi/tests/test_organize.py	95.29% <100.00%> (+0.14%)	⬆️
dandi/validate.py	97.87% <100.00%> (+0.14%)	⬆️
... and 11 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[yarikoptic]

• let's make at least _ses strictly following sub- if present
• according to https://github.com/dandi/dandi-cli/blob/HEAD/dandi/organize.py#L45 is still missing probe and obj.

[yarikoptic]

ideally there should be a single data structure describing entities etc we should have in the filename. In principle that potential_fields is the one but I guess can't be used here "as is", unless you enrich that structure with regexes for the values, in particular for extensions etc. Then this regex can be constructed programmatically and we would avoid duplication of this knowledge.

[jwodder]

@yarikoptic Can tis/slice/cell/probe/obj occur without a preceding ses, or is ses a prerequisite for them?

[yarikoptic]

according to potential_fields, ses is not mandatory, so other entities could occur without a preceding ses.

[jwodder]

Done: 55df999.

[yarikoptic]

ok, I dislike that we have duplication of specification ATM between organize and validate but ok -- let's RF it after.

Meanwhlie filed #1174 which found while test driving, and seems to work:

❯ dandi validate XCaMPgf/XCaMPgf_ANM471996_cell01.nwb
[NWBI.check_subject_species_form] XCaMPgf/XCaMPgf_ANM471996_cell01.nwb — Subject species 'mus musculus' should be in latin binomial form, e.g. 'Mus musculus' and 'Homo sapiens'
[NWBI.check_intracellular_electrode_cell_id_exists] XCaMPgf/XCaMPgf_ANM471996_cell01.nwb — Please include a unique cell_id associated with this IntracellularElectrode.
[DANDI.NON_DANDI_FILENAME] XCaMPgf/XCaMPgf_ANM471996_cell01.nwb — Filename does not conform to Dandi standard
[DANDI.NON_DANDI_FOLDERNAME] XCaMPgf/XCaMPgf_ANM471996_cell01.nwb — File is not in folder at root with subject name

although messages are a bit cryptic I guess.