Implement clean up strategies

[sujaypatil96]

This PR seeks to implement some of the repo clean up strategies discussed in issue #105. The major change in this PR is the integration of the sheet2linkml package as a dependency in the project, and the consequent removal of the vendor directory which previously housed sheet2linkml.

[turbomam]

I'm OK with the examples-for-discussion → examples directory name change, but I don't remember any discussions bout it. Didn't somebody from outside of the tools group create that directory or contribute to it? Is everyone involved OK with the name change?

[turbomam]

why are we renaming model/→ linkml_model? Is this something that we're going to start doing for most linkml projects that BBOPers contribute to?

[turbomam]

• I'm OK with the examples-for-discussion → examples directory name change, but I don't remember any discussions bout it. Didn't somebody from outside of the tools group create that directory or contribute to it? Is everyone involved OK with the name change?
• why are we renaming model/→ linkml_model? Is this something that we're going to start doing for most linkml projects that BBOPers contribute to?

[turbomam]

I haven't run this within this repo, but I have run it directly in cancerDHC/sheet2linkml, so I trust it

[gaurav]

I think the output file should be --output model/schema/crdch_model.yaml, unless we're moving the directory we store crdch_model.yaml in?

[turbomam]

BTW I'm glad you're taking the initiative to cleanup, @sujaypatil96

[sujaypatil96]

I'm OK with the examples-for-discussion → examples directory name change, but I don't remember any discussions bout it. Didn't somebody from outside of the tools group create that directory or contribute to it? Is everyone involved OK with the name change?

Hm, I think I did bring it up during one of our meetings, but there wasn't a particularly long discussion about it. That makes sense, I CC the people involved and check if they're okay with the name change.

Edit: There's a discussion thread for the above here.

why are we renaming model/→ linkml_model? Is this something that we're going to start doing for most linkml projects that BBOPers contribute to?

I though it'd be a good idea to rename it to linkml_model since in the same way that we have the crdch_model folder in the repo, which is pretty self explanatory as to what it houses, model was too generic to be semantically useful. Which is why I recommended a change to linkml_model so we know its contents.

[turbomam]

Edit: There's a discussion thread for the above here.

Thanks. We should check with @gaurav and make sure that we're not confusing the examples-for-discussion with examples that come from the example-data repo

[gaurav]

Might overlap with #125

[gaurav]

I would stick with model/ instead of linkml_model/ as the directory for the model -- partially because I think of linkml-model as being the only true "linkml model", but mostly because I think we should follow linkml-model-template in referring to this as the "model" directory. I think the best name for this directory might be "src", but if so, that change should be made in linkml-model-template before it propagates down to us, so that all LinkML models can share a similar directory structure, making it easy for LinkML users to jump between them.

[gaurav]

I think the output file should be --output model/schema/crdch_model.yaml, unless we're moving the directory we store crdch_model.yaml in?

[turbomam]

I don't have 3.7, so tox skipped that. 3.8 and 3.9 passed.

Maybe somebody can teach me how to install an old version of Python in a virtual environment for more thorough tox testing. I haven't been in the habit of installing old Python versions into my global environment.

Also, why does the 3.8 test bundle work on my computer but not within the GH action?