Requirements for um_fields_subset.py

[atteggiani]

Drafting some requirements for um_fields_subset.py to help out with refactoring / unit-testing / regression testing:

General requirements

Primary functionality: Remove fields from a UM file, generating an output UM file with only a subset of its initial fields in it.
Python version: >=3.10
Audience: Researchers / ACCESS model users
Mode of use: Offline. To be run manually by the user.
Interface type: CLI

Interface requirements

Inputs: Input file (positional argument) [UM file, required]
Outputs: Output file (output path provided by the --output option)
Options:

• -p, --prognostic-only: Include only STASH item codes of prognostic variables (section 0,33,34). Cannot be used together with --include or --exclude.
• --include: List of STASH item codes. Only these fields will be included in the output UM file. If not present in the input file, raise a warning. Cannot be used together with -p/--prognostic-only or --exclude.
• --exclude: List of STASH item codes. These will be the only fields excluded from the output UM file. If not present in the input file, raise a warning. Cannot be used together with -p/--prognostic-only or --include.
• -o, --output: output path (if not provided a default output path is produced)

Note:
Feel free to add specifications/options if I missed any relevant one.

[leoberhelman]

We would still want the section option correct? Allowing the user to get rid of sections. Or do we just want it to be by specific STASH codes given

[atteggiani]

We would still want the section option correct? Allowing the user to get rid of sections. Or do we just want it to be by specific STASH codes given

No, I think just the specific STASH codes will do.

[atteggiani]

Also look at #12 for additional requirements.