update min python to 3.9

[martenson]

otherwise this error occurs: Error: The version '3.7' with architecture 'x64' was not found for Ubuntu 24.04.

[mvdbeek]

I guess it's time to update the syntax too.

[martenson]

yeah, seems we have some linting debt, I'll update it

[mvdbeek]

We might have to go to 3.9 for pydantic ... 3.8 is also EOL already

[martenson]

@mvdbeek sounds reasonable, I'll bump it

[mvdbeek]

Then we need to pin pydantic 2. I think the this would work with pydantic 1 if we are able to install a newer pydantic by bumping the python version.

[mvdbeek]

(No strong preference either way)

[martenson]

what about both, pydantic2 pin and py39 as minimal?

[mvdbeek]

yeah, let's do it. we've supported 2 for quite some time, just hope there's no other breakage or incompatibility.

[martenson]

I forgot new lint will hop out with ne python 🤣

[martenson]

pydantic failing on missing fields, but afaik those two are Optional

=================================== FAILURES ===================================
_____________________________ test_idc_lint_valid ______________________________
tmp_path = PosixPath('/tmp/pytest-of-runner/pytest-0/test_idc_lint_valid0')
    def test_idc_lint_valid(tmp_path: Path):
        setup_mock_idc_dir(tmp_path)
        output_path = tmp_path / "output.yaml"
>       write_shed_install_conf(tmp_path / "data_managers.yml", output_path)
tests/test_idc_data_managers_to_tools.py:11: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
data_manager_conf_path = PosixPath('/tmp/pytest-of-runner/pytest-0/test_idc_lint_valid0/data_managers.yml')
output_path = PosixPath('/tmp/pytest-of-runner/pytest-0/test_idc_lint_valid0/output.yaml')
    def write_shed_install_conf(data_manager_conf_path: str, output_path: str) -> None:
        tools_yaml = build_shed_install_conf(data_manager_conf_path)
    
        # validate generated dict to ensure we're writing out valid file
>       RepositoryInstallTargets(**tools_yaml)
E       pydantic_core._pydantic_core.ValidationError: 2 validation errors for RepositoryInstallTargets
E       api_key
E         Field required [type=missing, input_value={'tools': [{'name': 'data...olshed.g2.bx.psu.edu'}]}, input_type=dict]
E           For further information visit https://errors.pydantic.dev/2.10/v/missing
E       galaxy_instance
E         Field required [type=missing, input_value={'tools': [{'name': 'data...olshed.g2.bx.psu.edu'}]}, input_type=dict]
E           For further information visit https://errors.pydantic.dev/2.10/v/missing

[mvdbeek]

that's the difference between pydantic 1 and 2. Optional[thing] means union of thing and None. If you want to allow omitting the value you need to provide a default,i.e Optional[thing] = None. That matches the signature logic of python functions.