Correct way to use the CLI

[gabriel-abrahao]

Hi all,

Basic question here. I'm wondering what is the intended way to use the CLI and the scripts contained in scripts. With a development version, I can just call e.g. python /path/to/cloned/climate-assessment/scripts/run_harm_inf.py --flags. However, the scripts folder is not installed when using pip. So, what was the original plan to install and use the CLI scripts? Are they supposed to only be called inside python, or is Anaconda the only supported way to use the CLI?

I ask because we are trying to avoid actual python code in the REMIND implementation, and stick to the end-user scripts as much as possible. Of course if that's not supposed to be supported, we can get some python in there, or I can do the work to make the CLI more "installable" if someone points me in the right direction.

Thanks anyway

[jkikstra]

Hi Gabriel, thanks for opening this question.

Note that at the moment there's no one "correct way to use the CLI". As we developed this for AR6, we do not currently provide full support for functionality, so there's a bit of "let's all try to contribute and keep it up-to-date when we can" and "if it works, great" :).
Indeed, scripts itself is mainly there to provide ways of how to use it, but the current idea was not to also include it in the package distribution (it rather should be scripts that make use of the package). As mentioned by Pascal in the thread you linked, you generally should be able to import from cli.py

Some use cases:

• I myself personally only use it directly, virtually always calling just run from Anaconda, like mentioned in the docs: https://climate-assessment.readthedocs.io/en/latest/user_guide.html#preparation-steps-before-running-from-command-line-interface-cli
• In another use case, developed by @phackstock, the development version is installed directly from GitHub (https://github.com/iiasa/climate-assessment.git), and then the run_worklow call is using from climate_assessment.cli import run_workflow.
• For a notebook example of the same thing, you can also check out https://github.com/iiasa/climate-assessment/blob/main/notebooks/run-example-magicc.ipynb