VPP001: Error: `tool.ruff.lint` cannot be validated by any definition

[DimitriPapadopoulos]

See:
https://learn.scientific-python.org/development/guides/repo-review/?repo=pypa%2Fsetuptools_scm&branch=main

The error message is unclear, I think the failure originates here:

[tool.ruff.lint.isort]
force-single-line = true
from-first = false
lines-between-types = 1
order-by-type = true

[henryiii]

Are one of those fairly new? We might need ti bum validate-pyproject-schema-store. Or the SchemaStore might need a new update (which the stuff team does pretty regularly).

[DimitriPapadopoulos]

None of these settings is very recent, the most recent is from-first from 2023 (astral-sh/ruff#8663).

In the error message, the isort key is almost empty:

    'isort':
      at least one of the following:
        - {}
        - {type: null}

Is that expected?

[henryiii]

Hard to check on my phone, is it in here: https://json.schemastore.org/ruff.json

[DimitriPapadopoulos]

It points to the IsortOptions definitions:

[henryiii]

It's mad about

lint.select = ["E", "F", "B", "U", "YTT", "C", "DTZ", "PYI", "PT", "I", "FURB", "RUF"]

I think because it has preview only selections in it.

[henryiii]

The issue is "U" is not a valid code. "UP" is the correct code. See pypa/setuptools-scm#1064.

[DimitriPapadopoulos]

What got me confused is that there were 2 error messages for the same issue;

• RF103: pyupgrade must be selected. Must select the pyupgrade UP checks.
• VPP001: Error: tool.ruff.lint cannot be validated by any definition

The error message could be more helpful (perhaps show the actual invalid code).

[henryiii]

For validation, that's handled by validate-pyproject, maybe the error message can be improved there (it's from fastjsonshema). For our end (sp-repo-review, that is), RF103: pyupgrade must be selected. Must select the pyupgrade UP checks. described exactly what was missing (no UP code).

[DimitriPapadopoulos]

Because RF103 had already been raised about the UP typo, I was misled into looking elsewhere for the VPP001 cause.

The same root cause should not result into different error messages, especially without enough details to clearly identify the root cause (at least a line number).

[fedorov]

I am experiencing the same issue here: https://github.com/ImagingDataCommons/idc-index/actions/runs/17810608557/job/50633246476?pr=174. I cannot figure it out, would appreciate any suggestions!

[jcfr]

@fedorov I am looking into the issue associated with idc-index 🚀

[fedorov]

Thank you JC!

Is there anything that can be done for those poor souls who do not have someone like you to come to the rescue?

For the record, I did try coding assistants - Copilot, Perplexity and Claude - and neither one was of any use in solving this. With my prompts, anyway...