Unified data mapping syntax #14
Description
Some data mappings approaches choose to code directly (usually in Python or Matlab) the mapping of native experimental data into IMAS IDS (imperative approach), whereas others are trying to express the mappings using some kind of markup language that describes relationship between native names and IDS paths (more like a declarative approach).
While both approaches have advantages and shortcomings, it may be advantageous in both case to anyway specify the mapping relationships (names + units) in a textual format, to serve for documentation as well as for Agentic AI purpose. Then the benefit of sharing a common language/syntax for expressing this information is obviously to improve the communication between different groups and data experts and foster collaboration, contributions and peer-review.
In this issue we can exchange ideas, mapping syntax snippets and links to better prepare and conduct this first topical track during the 1st Workshop on IMAS Data Mapping. Ideally, we could complete a review of syntaxes currently in use before the meeting, maybe even analyze pro & cons and how compact (and human friendly) these are. During the meeting we can finalize a draft unified syntax and port existing formats to it to verify it's adequacy to capture all exiting information. If this is successful we can plan common tooling around this format.