Made FSH files valid and some additional features #508
Conversation
…e for the base layer
…cardinality of extensions
…e to fix this in the current codebase.
…ints which where missing
|
Nice work @ArdonToonstra! I did not use GoFSH as I'm more knowledgeable with the profiles as they are than the output of the GoFSH conversion, so this seemed easier. Anyway, this is meant to use as a one-off script, next step will to manually adjust the result for the missing parts (already made a start). For now, I don't want to use the default IG organization. It requires careful consideration to see if we want to go that route, and if so, if there should be a different base and core IG. So I'll merge in your changes and start working from there on adjusting the result, with the aim of having nl-core profiles that are the same as before. I actually found a reasonable workflow for this;
|
| if el.min == "0": | ||
| # For slices that are already defined in the base profile, we must clear their | ||
| # cardinality to prevent them from being defined again in the core profile. | ||
| if el.slice_name and not el.slicing_path: |
There was a problem hiding this comment.
I think this is incorrect, as it removes the cardinality constraints from slices and extensions, which should be added here. Where is this resulting in unwanted output?
Various improvements to the zib2fsh.py script, making the output FSH valid and adding some missing features.
Placed it inside a default sushi config directory for easy IG building for validation purposes.
@pieter-edelman-nictiz -- I was wondering if you considered using GoFSH first, and then start scripting on the fsh files?