docs(DOD): 补 v2.2.6 + #144 这一轮的执行记录(四组证据按强度分开写) - #146
Merged
Merged
Conversation
上一版"最近一次执行"停在 2026-09-22 的 v2.2.4,之后两轮(#143/#144 + v2.2.6 发版)的取证只散在 PR body、提交信息和 issue #130 的关闭评论里 —— DOD 是"这轮到底验到哪一格"的唯一台账,缺它就等于 下一个人只能靠翻 PR 猜。补进来的按四组分开写,因为它们的强度不是一类: - 机器侧全量:#144 那轮 **2147 passed / 38 skipped / 0 failed,66.8s**(#143 那轮 2145/38/0,94.63s)。 - CI 侧:12 平台矩阵 + ruff + mypy 棘轮 + DCO 全绿,**Playwright E2E pass 9m2s**。这一格特意写明 "是 CI 跑的、不是我本机跑的",别跟 v2.2.4 那条"e2e 没在本机跑"混成同源证据。 - 真机(浏览器)侧:端口 7899、不加载模型、**显存全程 9638 MiB 恒定**、收工核实端口无监听 + 进程退出。 口径写死:它证明的是"事件链路在真浏览器里对、且没偷偷申请显存",**不是**模型加载/切换的显存曲线; 那格这轮零覆盖,与 v2.2.4 的"切换后 9489→3344 MiB"不是一类证据。 - 产物侧:RP 构建那份,wheel 28,381,933 B / 1065 条目、sdist 28,341,730 B / **1209 文件**(比 2.2.5 的 1208 多的那个正是新增测试文件 —— 差异能对上才说明数字不是抄的);sha256 **三方一致** (本地 `sha256sum -c` rc=0 == Release 的 SHA256SUMS == GitHub 自算 asset.digest);解包按包导入 `run_startup_selfcheck(enforce=True)` = `16/16/0 skipped=0 manifest_signed=true`。 另加:两个 PR 的取证脚本都改从**独立 worktree 的干净检出**跑 —— 主工作树那阵子挂着别人的未提交改动。 - 镜像侧:4 个标签推到 digest `dcedcf0d…`;**集群真拉一次仍未取证**(缺 `ghcr-pull` secret,要 带 `read:packages` 的 PAT,属 owner 动作)—— 这句写进去,免得下一个人把"run success"读成"能上线"。 没覆盖的格子照旧显式列出:字体菜单 / 断网首屏 / 侧栏换页竞态没重跑(上次取证 2026-09-19/20 的 v2.2.2 工作树)、模型加载与切换的显存曲线这轮没碰、`Setup.exe` 真机安装与 26 GB 分卷仍未验收。 type 是 `docs(DOD)`:release-please 侧 `docs` 落在 exclude-paths(`docs`)里,不会 mint 版本 —— 实测口径见 `docs/release-governance.md` §1 第 5 条。 复算:`tests/test_version_consistency.py` + `tests/test_no_jinja_in_inline_event_attrs.py` = 11 passed; `ruff check docs` 通过。`已发布最新 = v2.2.6` 一行未动。 Signed-off-by: ReSerendipity <zengyangc@outlook.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
这条只做一件事:把 2026-09-23 这一轮的取证写进
docs/DOD.md上一版"最近一次执行"停在 2026-09-22 的 v2.2.4;之后 #143/#144 与 v2.2.6 发版的证据只散在 PR body、提交信息和 issue #130 的关闭评论里。DOD 是"这轮到底验到哪一格"的唯一台账,缺它下一个人只能翻 PR 猜。
写进去的四组按强度分开,不混:机器侧全量(2147 passed / 38 skipped / 0 failed,66.8s)、CI 侧(12 平台矩阵 + ruff + mypy 棘轮 + DCO + Playwright E2E pass 9m2s,并标明是 CI 跑的)、真机浏览器侧(端口 7899、不加载模型、显存全程 9638 MiB 恒定、收工核实端口无监听与进程退出,且写明这一格不是模型显存曲线)、产物侧(RP 构建那份:1065 条目 / 1209 文件、sha256 三方一致、
run_startup_selfcheck(enforce=True)=16/16/0 skipped=0 manifest_signed=true)。另加一条防误读:镜像 4 个标签推到 digest
dcedcf0d…,但集群里真拉一次仍未取证(缺ghcr-pullsecret → 要带read:packages的 PAT,owner 动作)。不写这句,下一个人会把"run success"读成"能上线"。没覆盖的格子照旧显式列出:字体菜单 / 断网首屏 / 侧栏换页竞态没重跑;模型加载与切换的显存曲线这轮没碰;
Setup.exe真机安装与 26 GB 分卷仍未验收。复算:
test_version_consistency.py+test_no_jinja_in_inline_event_attrs.py= 11 passed;ruff check docs通过;已发布最新 = v2.2.6一行未动。type 是docs(DOD),落在exclude-paths: ["docs", "tests"]里,不会 mint 版本。