Missing some recent OBO releases; tracking file is out of date

[caufieldjh]

Describe the bug

Some recent OBO releases don't appear to have been transformed - e.g., CHEBI has a new version but the latest version on KG-OBO is 219, from November 2022. Under normal conditions, the KG-OBO build process produces a log file titled tracking.yaml. This keeps track of each new OBO ontology version and its corresponding graph transform. The file hasn't seen an update since 2022-11-26.

Version

107f982d96aed308912734c8d30eb560bd6ba6d6

Additional context

Will investigate for any Jenkins build errors.

[caufieldjh]

Aha, the build has been encountering an installation error:

00:01:58  Collecting sklearn
00:01:58    Downloading sklearn-0.0.post1.tar.gz (3.6 kB)
00:01:58    Preparing metadata (setup.py): started
00:01:59    Preparing metadata (setup.py): finished with status 'error'
00:01:59    error: subprocess-exited-with-error
00:01:59    
00:01:59    × python setup.py egg_info did not run successfully.
00:01:59    │ exit code: 1
00:01:59    ╰─> [18 lines of output]
00:01:59        The 'sklearn' PyPI package is deprecated, use 'scikit-learn'
00:01:59        rather than 'sklearn' for pip commands.
00:01:59        
00:01:59        Here is how to fix this error in the main use cases:
00:01:59        - use 'pip install scikit-learn' rather than 'pip install sklearn'
00:01:59        - replace 'sklearn' by 'scikit-learn' in your pip requirements files
00:01:59          (requirements.txt, setup.py, setup.cfg, Pipfile, etc ...)
00:01:59        - if the 'sklearn' package is used by one of your dependencies,
00:01:59          it would be great if you take some time to track which package uses
00:01:59          'sklearn' instead of 'scikit-learn' and report it to their issue tracker
00:01:59        - as a last resort, set the environment variable
00:01:59          SKLEARN_ALLOW_DEPRECATED_SKLEARN_PACKAGE_INSTALL=True to avoid this error
00:01:59        
00:01:59        More information is available at
00:01:59        https://github.com/scikit-learn/sklearn-pypi-package
00:01:59        
00:01:59        If the previous advice does not cover your use case, feel free to report it at
00:01:59        https://github.com/scikit-learn/sklearn-pypi-package/issues/new
00:01:59        [end of output]
00:01:59    
00:01:59    note: This error originates from a subprocess, and is likely not a problem with pip.
00:01:59  error: metadata-generation-failed
00:01:59  
00:01:59  × Encountered error while generating package metadata.
00:01:59  ╰─> See above for output.
00:01:59  
00:01:59  note: This is an issue with the package mentioned above, not pip.
00:01:59  hint: See above for details.
00:01:59  WARNING: You are using pip version 22.0.4; however, version 22.3.1 is available.
00:01:59  You should consider upgrading via the '/var/lib/jenkins/workspace/knowledge-graph-hub_kg-obo_main/gitrepo/venv/bin/python3.9 -m pip install --upgrade pip' command.

'scikit-learn' is coming in from Ensmallen Embiggen, but that uses the "correct" package name so this may be attempting to install an older version.

[caufieldjh]

The build is attempting to install:

ensmallen-0.8.36-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
embiggen-0.11.42.tar.gz

and those are both the most recent versions, and scikit-learn is also installed successfully. So what else has that dependency and is specifying it as sklearn?

[caufieldjh]

Resolved - ontologies are being updated, as is tracking file.