Skip to content

docs: reconcile merged qualification and contributor checkpoints - #5235

Open
LIHUA919 wants to merge 1 commit into
loopx-project:mainfrom
LIHUA919:codex/history-checkpoint-cleanup
Open

LIHUA919 wants to merge 1 commit into
loopx-project:mainfrom
LIHUA919:codex/history-checkpoint-cleanup

Conversation

@LIHUA919

Copy link
Copy Markdown
Contributor

Goal and result

Reconcile the shared-authority frontier and contributor board with current merged and reviewed work. The old entries advertised another UTF-8 guard and migration design work, and treated the merged SQLite read-proof optimization as an open implementation PR.

Changes

Validation and boundaries

@huangruiteng huangruiteng left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

REQUEST_CHANGES — [P2] GH-A05 把仍有当前 exact-head 阻断的 #4915 写成“已批准、只等维护者合并”,这会误导下一位贡献者和维护者。

精确 head:3d45c1b2eaea6071fa48c3b6df5dae27390fd9b3;比较基线:440a19aab2b69dc23748bb734fa58f2a53e2872f。本人检查三个文档的完整差异与其引用的当前状态,未查询 GitHub CI。

动机

本次按 contributor board 的准入与状态规则 和 shared-authority RFC 整理已合并代码、仍开放的 PR 与尚未满足的资格门槛,避免重复实现。方向合理:合并了 #4931 不等于 SQLite D2 完成;Windows reporter 验证和默认迁移也各有自己的关闭条件。

改动思路

复用现有 GH-A01/A03/A05/R5A 行和双语 RFC checkpoint,不新建路线图、权限或第二套任务。#5160 的 UTF-8 guard、#4397/#5162 的 canonical regression、#4328/#4931 的 D2 资格被放回既有 owning boundary。

具体改动

双语 RFC 正确区分 #4931 已合并与容量、恢复、消费者延迟、支持 runtime 和十天 soak 未完成,并保留 9 月 24 日“当时仍在 review”的历史语境。board 的 GH-A01/A03/R5A 也保留 reporter、精确 head 和原有资格缺口。

唯一阻断在 docs/development/contributor-tasks.md:108(GH-A05): #4915 当前仍是 OPEN,head 为 1bd8bfa8b5690f02aea752129ac4adc5866109e4,最新 正式 review 是 CHANGES_REQUESTED,而非 APPROVED。该 review 给出真实 project register 的 base 成功/head 失败:双默认 runtime 并存时,source guard 丢失已经提供的 explicit registry/effective root。不是因为非自身 PR 的红 CI 请求修改。

最小修复只需把这一行改为等待当前 head 的修复与重新验证/评审,然后再进入合并与 issue closeout;保留 #4915/#4800 的 canonical anchor,不另起 migration Todo,也不在这里改迁移实现。

对主干的风险

36 项 contributor-board/RFC-status 检查、uv run --extra test python examples/docs-governance-smoke.py、精确三文件的 public-boundary scan 与 git diff --check 均通过。但这些测试验证结构与引用,不会证明远端 review 结论;上述当前 head 的反例仍成立。错误状态会让维护者跳过修复,也可能让贡献者以为剩余工作只是 merge。

未来向前的 bounded refine 已体现在替换旧行、复用既有 checkpoint;更大的自动 GitHub→board 同步没有本次同因需求,不应为改一行引入新状态源。无 runtime、权限、frontstage 首屏或行为变更。

我的整体评价

REQUEST_CHANGES:其余 reconcile 有实际价值,但当前状态准确性是本次交付本身,不能因结构测试全绿忽略这个 false approval。修正 GH-A05 后,重新读 #4915 当前 exact head 及正式 review,再跑 board/RFC checks 即可。未要求本 PR 修复其他 PR 的代码、未合并或升级。

English verdict: REQUEST_CHANGES - 3d45c1b; GH-A05 incorrectly declares #4915 approved although its current head has an actionable CHANGES_REQUESTED review. Correct the one current-status row and recheck the exact-head review; other reconciliation and native docs checks are valid.

@huangruiteng

Copy link
Copy Markdown
Collaborator

按 contributor board 的“已有实现复用+当前状态准入”框架,3d45c1b2eaea6071fa48c3b6df5dae27390fd9b3 的分项结论:

  • 已合并代码与 D2/Windows reporter/native regression 资格分开,未把 perf(authority): reuse owned state during SQLite and archive replay #4931 合并冒充 D2 完成。
  • 不新建 migration 或 SQLite qualification 路线,沿用既有 canonical anchors,符合压缩而不是追加的要求。
  • GH-A05 的 approval 状态与 #4915 当前正式 review 冲突,必须仅修这条现状陈述。当前 review 的阻断来自真实 CLI 回归,不是无关红 CI。
  • docs governance 和 36 项结构 checks 通过,但不证明 GitHub 当前 review 结论。

结论 REQUEST_CHANGES;最小修复与证据详见本次 exact-head review。

本次 exact-head 完整 review

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants