Validate notebooks once per fetch or save

[kevin-bates]

Some prior network analyses found that each fetch and save of a notebook file resulted in double validation, presumably so the server could set a helpful message on the notebook model returned to the user since the validation performed within nbformat's read or write methods did not return any indication of a validation error.

nbformat was recently updated to enable applications to ask for validation error status on the initial read/write calls, thereby enabling notebook to skip its second call to validate the notebook. This pull request leverages those updated signatures now available in nbformat 5.2.0.

The existing validate_notebook_model(), however, will still call nbformat.validate() if a None validation_error parameter is provided. This was done in order to preserve existing functionality from potential calls by server extensions. The message generation code is shared by both by raising the exception present in the validation_error dictionary.

(The CI failures do not appear to be related to these changes.)

[codecov-commenter]

Codecov Report

Merging #724 (34edc9d) into main (5d962d7) will increase coverage by 0.01%.
The diff coverage is 92.85%.

@@            Coverage Diff             @@
##             main     #724      +/-   ##
==========================================
+ Coverage   70.95%   70.96%   +0.01%     
==========================================
  Files          62       62              
  Lines        7515     7523       +8     
  Branches     1190     1192       +2     
==========================================
+ Hits         5332     5339       +7     
- Misses       1826     1832       +6     
+ Partials      357      352       -5     

Impacted Files	Coverage Δ
jupyter_server/services/contents/fileio.py	75.52% <80.00%> (ø)
jupyter_server/services/contents/filemanager.py	71.73% <100.00%> (+1.00%)	⬆️
jupyter_server/services/contents/manager.py	88.60% <100.00%> (+0.41%)	⬆️
jupyter_server/services/kernels/kernelmanager.py	80.69% <0.00%> (-1.90%)	⬇️

---

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 5d962d7...34edc9d. Read the comment docs.

[blink1073]

Thanks! Can you please install and run pre-commit on all the files?

pip install pre-commit
pre-commit install
pre-commit run --all-files

[kevin-bates]

Thanks Steve, that was helpful.

[blink1073]

Kicking CI to pick up fixes