Add NewareNDA extractor entry #78
Conversation
| - method: cli | ||
| command: NewareNDA-cli.py {{ input_path }} --format csv {{ output_path }} |
There was a problem hiding this comment.
Hm, what is the python object generated from NewareNDA.read() below? Is it a dict, or some other object? I think we should probably have the analogous CLI usage (json?) for the python usage for now, instead of having 10 equivalent ones with an output format.
Also, we have output_type in the template: https://marda-alliance.github.io/metadata_extractors_schema/main/mme_schema/UsageTemplate.html
There was a problem hiding this comment.
It returns a dataframe at the moment, the CLI then just calls the various df.to_<x>() functions that pandas has.
output_type requires a MaRDA filetype itself right? I think it's also single-valued (at the moment) so not sure how useful it is here
There was a problem hiding this comment.
OK, then we probably need to mention pandas somewhere. I'll think about how to add that into schema (also for the xarray requirement in yadg) in another PR.
And yes, we should probably add those as FileTypes into the registry anyway.
|
This needs to be remade at the yard |
This PR adds the NewareNDA Python package as an extractor of our
neware-ndafile type.Perhaps of interest to @d-cogswell, the author of the package (feel free to suggest any changes).
In terms of implementation, I have added several very similar CLI invocations for each of the support output file formats, which could maybe be merged/templated after a future schema update.
We're also missing example files for the
neware-ndafile type (which is in fact two types,.ndaand.ndax-- could also be separated...). There's an open PR and issue about adding test data at Solid-Energy-Systems/NewareNDA#48 that we could poach from to set up our own validation pipelines here.