aboutcode-org / aboutcode-org/scancode-toolkit

Validate all license-digger licenses are correctly detected

未关闭
#4,141 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
主要语言
Python
星标
2.6k
派生
791
平均合并
1 天 12 小时
30 天内合并 PR
5

描述

We should add as test files most of the files from KDE's license-digger originally by @cordlandwehr @kossebau and team.
- The repo is at https://invent.kde.org/sdk/licensedigger with a mirror at https://github.com/KDE/licensedigger/
- Attached is a subset of the repo with files we should test: license text, templates and test suites. This is a nice data set as there is a lot of manual curation work that went into to review all and every license headers in the many project of KDE. See [licensedigger-master-cc4b24d3fb67afa8fb0a9ef61210588958eaf0f5.zip](https://github.com/user-attachments/files/18749172/licensedigger-master-cc4b24d3fb67afa8fb0a9ef61210588958eaf0f5.zip)

I am fairly confident that we do detect all these, but we need to verify this, and we need highly accurate detection.

- Any license file text that is detected exactly by a matcher as a whole file or fragment should not be added as test, as this would be redundant.
- When a license text is detected approximately, this would be a candidate for rule addition.

贡献指南

打开贡献指南

评估

这个 Issue 还没有评估数据。

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。