ml-evs
commented
4 years ago :wave:
Just to add that we over at Materials-Consortia/optimade-python-tools are currently working on adding tutorial-style documentation for how this might be achieved, I'd expect this to be achieved by the end of next week once we have some other changes out of the way.
In the meantime, I shared some of our tutorial/overview videos from the recent workshop with @volkerblum, hopefully these will also soon be made available on YouTube by the organisers.
This is a discussion thread in order to summarize activities related to supporting the OPTIMADE API, which is a generic API allowing the exchange of data between different, potentially disparate materials databases in a single, documented form.
A link to the broader effort is here: https://www.optimade.org/
A 1.0 version of the OPTIMADE API is expected within 2-3 weeks at the time of writing.
What we think will be needed is:
create a Django layer that takes OPTIMADE queries, ideally, using the OPTIMADE python tools package (this will enable parsers etc): https://github.com/Materials-Consortia/optimade-python-tools
we may need a way of adding a second web server on top to support the URLs associated with two different APIs (our own MatD3 REST API and the OPTIMADE API). The web server is already implemented in https://github.com/Materials-Consortia/optimade-python-tools ... supporting it may just be a case of customizing this existing work
need to make a decision of how to house the additional fields required for the OPTIMADE schema - either in our existing SQL schema or in a separate noSQL database altogether. The former seems more practical, in that it can be done incrementally, to me (VB)
This summarizes the discussion so far.
@ml-evs or @CasperWA are involved in this discussion and may also be able to help with questions.