fix(license): declare Apache-2.0 everywhere, matching the LICENSE file - #577
Merged
Conversation
pyproject.toml, build/package/nfpm.yaml, the Homebrew formula, the Claude Code plugin manifest and the README. So v0.79.0 onwards shipped a wheel telling PyPI one licence while the LICENSE file inside the same distribution said another, and the Chocolatey package pointed its licenseUrl at the Apache text under an MIT declaration. Nothing caught it because every file was individually valid -- the drift only exists between them. tests/test_license_consistency.py now pins all five declaration sites to the LICENSE file, including a catch-all that fails if any of them says MIT again. Verified on the real artifact, not just the source: `uv build --wheel` now emits `License-Expression: Apache-2.0` and `License-File: LICENSE`. No code change.
Devin review of #577. Both turned out to need documentation rather than a fix, but one needed checking rather than assuming. WinGet: `Keboola.KeboolaCLI` DOES exist in microsoft/winget-pkgs and declares 'MIT License' -- but it is the legacy Go CLI from keboola/keboola-as-code, a different product whose own LICENSE is MIT. Our id is `Keboola.KeboolaCLI2`, which has never been submitted (the winget job has never succeeded). So there is no kbagent manifest carrying a wrong licence today. The forward risk is real though: `wingetcreate update` only bumps version and installer URL, so the FIRST successful submission has to set License: Apache-2.0 itself. Noted in the workflow step, next to the code someone will edit when they fix it. Chocolatey: the nuspec has no SPDX field, only `licenseUrl`, which points at the repository's LICENSE and therefore self-updated when that file changed -- which is why this channel never went stale. Pinned by a test so it keeps pointing there rather than at some hard-coded licence page.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
pyproject.toml, build/package/nfpm.yaml, the Homebrew formula, the Claude Code
plugin manifest and the README. So v0.79.0 onwards shipped a wheel telling PyPI
one licence while the LICENSE file inside the same distribution said another,
and the Chocolatey package pointed its licenseUrl at the Apache text under an
MIT declaration.
Nothing caught it because every file was individually valid -- the drift only
exists between them. tests/test_license_consistency.py now pins all five
declaration sites to the LICENSE file, including a catch-all that fails if any
of them says MIT again.
Verified on the real artifact, not just the source:
uv build --wheelnowemits
License-Expression: Apache-2.0andLicense-File: LICENSE.No code change.