derived-census -check and pilot-diff -check gate no pull request or merge-queue run
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 4/5
- 预计耗时
- 1-2 天
- 新手友好度
- 35/100
- Issue 类型
- 功能
- 描述清晰度
- 基本清楚
- 活跃度
- 活跃
- 领域
- ci-cd, testing-qa
调研方向
Start with .github/workflows/pr.yml, job apijson-differential, and .github/workflows/oracle-reproduction.yml, which show how the pilot tools are provisioned and where pilot-diff -check runs today. Then read the derived-census command under tools/cmd to see what -check and -require-exporter read. Done means a pull-request-tier job fails when pilot-diff or derived-census baselines drift, with the path filter and JVM version settled as the issue describes.
由索引模型根据 Issue 内容生成。
描述
What happened, and what you expected instead
Two committed baselines are checked by no workflow that gates a pull request, so a change that moves either one passes the push tier and the merge queue and develop stops reproducing its own records. A pull request that moves a generated baseline should fail before it merges.
| Check | Where it runs today |
|---|---|
go run -C tools ./cmd/pilot-diff -check |
.github/workflows/oracle-reproduction.yml only (scheduled daily, and on dispatch) |
go run -C tools ./cmd/derived-census -check |
no workflow; .github/workflows/rebaseline.yml runs derived-census -update only |
go run -C tools ./cmd/pilot-apijson-diff -check |
.github/workflows/pr.yml, job apijson-differential, when the path filter matches — for comparison |
This is how #2223 landed: a name-resolution change moved one specification example in docs/project/pilot-differential-baseline.json and sixteen records in internal/semantic/metamodel/derived_census_records.tsv, and neither movement was seen until someone ran the checks by hand.
Measured cost
On an 8-core Linux machine with the corpora and pilot tools already provisioned (pin 2026-09, Java 21):
| Command | Wall time |
|---|---|
pilot-diff -check |
about 2.5 min |
derived-census -check -require-exporter |
about 14 min (the pilot export dominates) |
pilot-apijson-diff -check |
about 16 min |
derived-census -check without -require-exporter re-measures our side only and reads the pilot column from the committed baseline; it needs no Java and is the cheaper form, not timed here. Provisioning (the corpus downloads, the two validators and the JSON exporter) is extra and is already done by the apijson-differential job for its own check.
Options
- Run
pilot-diff -checkandderived-census -checkin the merge-queue tier ofpr.yml, path-filtered likeapijson-differential(the semantic, check and export layers, the tools module, the baselines). - Add both to the existing
apijson-differentialjob, which already provisions the same inputs, so the export cost is paid once. - At minimum, add
derived-census -check -require-exportertooracle-reproduction.yml, so the nightly run reports the census as it reports the differential.
The pilot column of the census also depends on the JVM for one declared cell (#2224); a gate that re-exports should run after that is settled, or on the Java version the baseline was recorded under.
- 主要语言
- Go
- 星标
- 59
- 派生
- 20
- 平均合并
- 10 小时 23 分钟
- 30 天内合并 PR
- 897
环境准备
- 提供 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
Open-MBEE/OpenSysML 的其他 Issue
-
难度 2/5 1-3 小时 新手友好度 76/100
维护者通常 1 天内回复
-
难度 5/5 一周以上 新手友好度 35/100
维护者通常 1 天内回复
-
难度 5/5 一周以上 新手友好度 35/100
维护者通常 1 天内回复
-
难度 4/5 3-5 天 新手友好度 55/100
维护者通常 1 天内回复
-
难度 5/5 一周以上 新手友好度 28/100
维护者通常 1 天内回复
查看 Open-MBEE/OpenSysML 的全部 Issue
相似的 Issue
-
难度 1/5 1 小时以内 新手友好度 78/100
OpenSlides/openslides-go#342 ·
维护者通常 4 天内回复
-
bug
难度 2/5 1-3 小时 新手友好度 77/100
-
docs: link to the CLA in InditexTech/foss instead of keeping a local CLA.md可能已有人在做 @amagan 今天认领。 未关闭documentation
难度 1/5 1-3 小时 新手友好度 88/100
InditexTech/redkey-operator#129 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 74/100
维护者通常 1 天内回复
-
难度 1/5 1 小时以内 新手友好度 88/100