Repository navigation
测试：在完整记录边界采集替换会话快照 - #91
Merged
JavaZeroo merged 4 commits intoOct 26, 2026
Merged
Conversation
Follow up the passive readiness experiment for #87; hold both replacement consumers at actual LF boundaries without weakening the original workflow oracle.
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
This was referenced Oct 26, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
原八会话替换检查点只等待编号前缀，可能把一次 8192 B 消费结束处的正常半行交给精确 oracle。改为默认关闭的有界记录检查点：各替换会话读至实际 LF 后保持，两个会话均就绪才采集并释放；关闭解除保持，关闭中的窗口拒绝重新启用。原 32000 条/2048000 B、4096 B 分块、20 ms 间隔、五轮替换、70/90 秒期限及 Python 文字/身份/回收断言不变。
对应 #87，目标分支为实验
fix/conpty-backend-r47，不选择正式产品核心/后端。说明与白名单证据见 机制合同 和 固定验证。replacement output not active，作为被否定方案保留c662011，candidate .5 / official 1.25 flags243a0561c，实际 checkout3fdd147历史两份失败缺失的 native 后续不追溯补造。新诊断 collected trace 超过128KiB，但所需补齐字节存在于保存前缀，不宣称全程运输已捕获。原件与本机路径留在本地。关闭 guard 红例未独立保留 EXE 哈希，公开字段为 null，源码/同一测试/build log/失败日志保留。更早
f89b0ac的57/57属于另一源码，不能替代最终结果。候选 CI 增加 Windows/Linux 的
workflow_checkpoint_test。#52/#63/#47 与 Draft #50 继续开放，真实 CF/IME、性能、WindTerm 等价及完整 R2 未由本项验收。Standards
终审无阻塞，可提交。公开 JSON 与 index 字节一致；最终源码、二进制、日志及两轮原件哈希匹配。56/57、402.66 秒、8 skip、Vim59 与旧版57/57分列准确；来源、隐私、未知和验收边界保留，CI精确选两测试。无新增判断性异味，原空行谓词重复为非阻塞建议；未重跑GUI/全套。
Spec
无阻塞发现，可限定关闭 #87 的合成工作流记录边界采样问题。原负载、期限、精确文本及身份断言保留；最终 c662011 两轮原件哈希、原 oracle 离线复算一致，各五轮 F4/F5 均仍活跃，Job0、desktop 回收。受控因果与正式运行分列；56/57、Vim59、8 skip、历史缺口及诊断截断如实保留。不覆盖 #52/#63/#47 或 R2。
两轴均无阻塞；非阻塞建议仅 Standards 一项，不改变限定验收范围。