Matthew Evans
Matthew Evans
I've also just run into this when trying to use `retrieve_queryable_properties` inside the entry mapper code (which means it also gets used for the enum-heavy `LinksResource` model...
For odbx, I have an extended `MongoEntryCollection` class that takes a pymongo index model and creates the index if not present (see below). Currently I only do this for `id`...
Looks like the Python package just validates against https://ucum.nlm.nih.gov/ucum-service.html#validation, which we could also do directly (e.g. https://ucum.nlm.nih.gov/ucum-service/v1/isValidUCUM/Ao for Angstrom)
I'm happy with all the wording here now, modulo processed vs derived etc (but happy to bow to consensus). I guess this is one of our only recent scientific additions...
Just a technical comment that [GridFS](https://pymongo.readthedocs.io/en/stable/api/gridfs/index.html#module-gridfs) would be the canonical MongoDB way of handling binary blobs (see also https://docs.mongodb.com/manual/core/gridfs/). As I missed the discussions I don't really understand the use...
This is drifting a bit from the original caching intention of this issue, but I quite like the idea of mapping filters (or at least providing classes to map filters)....
Where should the links go, to the section in the optimade.rst on GH?
Also I think we should do this before we make our next PR to the spec repo, which I guess makes this highish priority...
Hmm, maybe we should wait until that is all set up then!