Add `curie_map` to the model

gouttegd commented 4 months ago

Resolves #225

[x] docs/ have been added/updated if necessary
[x] make test has been run locally
[ ] tests have been added/updated (not applicable – nothing new to test, the CURIE map has existed for a while, this PR just formalises it directly into the model)
[x] CHANGELOG.md has been updated.

If you are proposing a change to the SSSOM metadata model, you must

[x] provide a full, working and valid example in examples/ (no need for a new example – all existing examples already have a curie_map)
[x] provide a link to the related GitHub issue in the see_also field of the linkml model
[ ] provide a link to a valid example in the see_also field of the linkml model (no need, see above)
[x] run SSSOM-Py test suite against the updated model

This PR makes the curie_map a bona fide slot of the MappingSet class, that exists independently of the serialisation format.

It builds on top of #349 by explaining in the text version of the spec what the curie_map slot is for, and how it should be dealt with in the SSSOM/TSV serialisation.

matentzn commented 3 months ago

Awesome thank you!!!

gouttegd commented 3 months ago

Sorry to repeat this in every PR

No need to be sorry. :) However:

Re-generating and committing all derived files upon every PR is, I believe, a very bad idea.

It makes even the slightest PR much larger than it needs to be, and therefore more cumbersome to review.
It pollutes the history of the project with large changes that do not mean anything (only the changes to the source schema are actually meaningful).
It greatly increases the likelihood of merge conflicts when there are several PRs in the works.

It would be much better in my opinion to leave the auto-generated files untouched most of the times, and only commit them once in a while (typically shortly before a release).

If you insist on having the derived files re-generated upon every PR, then it‘s something that should be mentioned in the CONTRIBUTING.md file, which for now has no trace at all of such of requirement – instead it’s full of useless, generic guidelines that were mindlessly copied over from another project.

And since there is a PR template, that requirement should also be mentioned in it.

gouttegd commented 3 months ago

I‘ll fix the conflict here once #375 will have been (re-)approved – no point in fixing it now since merging #375 will (again) cause a conflict in the changelog.

mapping-commons / sssom

Add `curie_map` to the model #376