Matthew Evans
Matthew Evans
Hi @JPBergsma, I've just changed the base branch of this PR and others to the new 0.25.x branch until we can port things over. I would suggest you keep working...
Looks great so far @JPBergsma! > It looks like the dropping of the python 3.8 typing was reverted the 0.25.x branch but it is still in this branch. So I...
> Thinking more about this, I think we should instead keep the repo as it is now, but instead set an environment variable in Heroku to locate the desired config...
I'm looking at the heroku now (Leo gave me access) and it looks like it just uses `Pipfile` and `Procfile`, not the docker. I'll investigate further... (if anyone else wants...
> @ml-evs are you OK with the heroku limitations? I'd still propose to try zeit.co. I haven't used either so happy to defer to someone who has. Either way, the...
> @merkys @ml-evs What do you think - should we still add this to "core OPTIMADE" as decided in the web meet. Or, now that we are hopefully somewhat close...
Great! Thanks for this @rartino -- I definitely stalled in my attempts to do the same thing. I will try to migrate https://github.com/Materials-Consortia/optimade-stability-namespace in the same direction.
> > The introduction of the SMILES data type will most likely also require some changes similar to those introduced in PR #444 for the timestamp data type. > >...
Related to #915.
This sounds related to #520?