Skip to content

Protect C1 collection diagnostics in quota presentation tests - #9002

Merged
huangruiteng merged 1 commit into
mainfrom
codex/c1-collection-failure-fixture-20261118t072236
Nov 18, 2026
Merged

huangruiteng merged 1 commit into
mainfrom
codex/c1-collection-failure-fixture-20261118t072236

Conversation

@loopx-agent

Copy link
Copy Markdown
Collaborator

Goal and delivered outcome

C1 returns the original quota diagnostic when collection cannot produce a complete work contract. The existing regression failed before exercising this behavior because its patch targeted the retired envelope builder. The test now watches the shared Effect Runtime boundary, compares the entire diagnostic, and retains the rejection of a retired protocol packet. Finding on the settlement regression review

The diff updates one test file. The current production path returns the same collection failure for ordinary quota, both compatibility parameters, and context delivery Off.

Validation

  • Before the change, the projection regression failed on the removed attribute. On the submitted commit, 86 related tests pass.
  • The actual CLI returns the same collection diagnostic for an unregistered synthetic Goal across all four entry modes. Each response exits 1 with zero admission receipts and zero debit records.
  • Configured Ruff, changed-file lint, mypy, the paired RFC check, the public-boundary scan and signed diff checks pass. All five standard premerge checks selected for this diff pass and the exact change-quality receipt is valid.

Written by OpenAI Codex.

Signed-off-by: LoopX Agent <337587101+loopx-agent@users.noreply.github.com>

@loopx-agent loopx-agent left a comment

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

Reviewer: model_agent · gpt-6.2-sol · OpenAI · runtime_reported · reasoning_effort=max

Approval conclusion (author-owned PR; GitHub blocks formal self-approval)

评审提交 ee5ac3ec2220d04454e33ff1b6a971a48dbe54d2,基线 f9bb5e3cf9185de77a2f7580d64eaf9d32f46eff。精确差异

动机 / Motivation

维护 C1 的开发者需要在 quota 收集失败后收到原始拒绝原因,并在预合并时发现诊断丢失。原测试在安装已退役的 envelope builder mock 时先报 AttributeError,尚未验证失败会如何返回。修订后的测试监视现行 Effect Runtime 边界,比较完整失败对象,并要求旧输入仍被拒绝。此次范围是测试修复,生产投影、CLI 参数和记账协议沿用现行实现。

改动思路 / Approach

quota should-run 的收集结果进入 _project_quota_cli_payload。投影先调用现有的退役格式检查。对缺少交互契约的收集失败,直接返回原字段。测试在 Effect Runtime 入口设失败哨兵,两个参数化分支都经过现行投影,分别断言完整对象和原样输入。现有投影

具体改动 / Changes

单文件 +12/-13。对照现行拒绝约定。复验 原始拒绝诊断 时,普通入口、两条兼容参数和 Context Off 在基线与候选的八次隔离 CLI 调用均退出 1,返回同一字段集与同一 quota_collection_failed。比较只去掉各次自动生成的观察事件 ID 和时间戳,准入与扣额均为零。复验 兼容输入拒绝 时,带 protocol_action_packet 的输入仍从原验证器抛出 EffectRuntimeRejected,两个测试分支都核对输入未被改写。回归测试 退役格式验证器

对主干的风险 / Risks

基线不带旧格式的分支在挂载已退役 mock 时失败,带旧格式的分支仍拒绝。候选整份测试 46 项通过。配置内 Ruff、mypy、签名及标准本地预合并五项直接检查通过。隔离 CLI 使用不存在的合成 Goal,原注册表字节不变,持久化只见跳过观察事件。

我的整体评价 / Verdict

同意这项测试修复。它修好原有回归并保留完整诊断比较及退役边界;命令行比较确认生产路径行为一致,避免把测试修复误作运行时功能变更。

English verdict: APPROVE. The exact head restores the collection-failure test at the current runtime boundary and preserves the complete diagnostic and retired-input rejection.

@huangruiteng
huangruiteng merged commit a067abd into main Nov 18, 2026
30 of 36 checks passed
@huangruiteng
huangruiteng deleted the codex/c1-collection-failure-fixture-20261118t072236 branch November 18, 2026 11:04
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