Ivan Ruiz Manuel
Ivan Ruiz Manuel
Alright. With that, I guess we can use the design in this document (with the small change above) as the goal for all `pydantic` PRs until the next release. If...
@brynpickering I did not mean to say that `data:` would replace `math.dims`. I think specifying `dims` metadata there is the right call. You are also right in saying that if...
Solved by #783
For posterity, I'll try to summarise the Status-Quo as well as design discussions between @brynpickering and me here. Below is a comparison of the **model.py** file in main (as of...
@brynpickering I aggree with Stefan. What order should we follow, given #783? I think this one could be merged first before continuing there.
@brynpickering is it possible to add this feature without making our tables functionality super complex? In all honesty I think we'd gain more from supporting better defined formats (like [parquet](https://github.com/calliope-project/calliope/issues/694))...
@brynpickering wouldn't it be possible if the `MODEL_SCHEMA` was constructed inside the `Model` class instead of a global? We already have the `update_model_schema` function, meaning that it is def. possible...
You are right on that! Sorry, the `MODEL_SCHEMA` bit was wrong on my end. But yeah, the general idea is to maybe have some step were we combine stuff into...
@brynpickering should this be fixed before #551 is approved?
- First comment: I agree. Storing the overrides would be even better for repeatability, and would make debugging easier! I have a design in mind, I will just call theses...