Skip to content

docs: record verified QA release and remaining OCI blocker - #113

Merged
Kitkitkittt merged 8 commits into
mainfrom
codex/98-price-unit-workspace-release
Oct 10, 2026
Merged

Kitkitkittt merged 8 commits into
mainfrom
codex/98-price-unit-workspace-release

Conversation

@Kitkitkittt

Copy link
Copy Markdown
Owner

Refs #111 and #112. Record exact merged CI, production frontend identity, live navigation/warm-reload evidence and immutable ARM64 image digest. Backend rollout remains blocked on Tailscale SSH approval; financial/provider acceptance is not misreported. Remove a test-only object-identity assertion while retaining behavioral TTM value and rejection checks. Focused financial-service tests: 38 passed. No runtime changes; preserve unrelated glossary/Matrix work.

Refs #98, #100, #101, #102, #103, #104, #105, #106, #107, #108, #109, #110, #111. Preserve unrelated glossary and Matrix work; record local verification and deployment limits.
Refs #101 and #111. Preserve requested TTM identity and null metrics for missing quarterly data or ordinary calculation failures. Keep cancellation semantics; verify HTTPS trust in deployment smoke. Full release gate: 157 frontend suites/1044 tests, 1610 backend passes and 4 skips.
@vercel

vercel Bot commented Oct 9, 2026 •

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
vnibb-web Ready Ready Preview Oct 10, 2026 3:49am UTC

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 9, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-09T14:21:29.787723Z 7bb7b37 PR opened
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 7bb7b37a6c

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment on lines 99 to 101
if metric_key in PERCENT_METRIC_LIMITS:
if 0 < abs(numeric) <= 1:
numeric *= 100

limit = PERCENT_METRIC_LIMITS[metric_key]
while abs(numeric) > limit and abs(numeric / 100) <= limit:
numeric /= 100

if abs(numeric) > limit:
if abs(numeric) > PERCENT_METRIC_LIMITS[metric_key]:
return None

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P1 Badge Do not expose unscaled provider fractions as percentages

When a provider supplies fractional ratio values such as roe=0.25, this now returns 0.25 for a metric declared and rendered as a percentage, producing 0.25% instead of 25%. The persistence path stores ratio.roe, roa, roic, and margin values directly in data_pipeline.py without converting or tagging their scale, and this function's own comment acknowledges that provider paths emit fractions. Normalize these values at the write boundary or reject values whose scale is unconfirmed rather than presenting the stored fraction as a canonical percentage.

Useful? React with 👍 / 👎.

@Kitkitkittt

Copy link
Copy Markdown
Owner Author

Final bounded Standards and Spec reviews found zero findings in be1c186/7bb7b37. Focused financial-service tests passed38; runtime unchanged. Vercel preview passed. No new GitHub CI run was observed for this follow-up, so this PR is not merged on the assumption of green CI. Deployed runtime remains main0e5dfdd with verified CI; OCI rollout/new-backend financial acceptance remain blocked by human Tailscale approval. Durable production/image evidence: #112 (comment)

@Kitkitkittt

Copy link
Copy Markdown
Owner Author

User renewed finish/commit/redeploy authorization. Synchronized branch with merged main in isolated worktree; three squash-ancestry conflicts retain verified follow-up docs and remove only the incidental object-identity assertion. Commit197ca5331978781fd6989a9f0fe03e89d3b95063; focused financial-service tests38 passed. CI now started: https://github.com/Kohnnn/vnibb/actions/runs/38021759796. Unrelated staged glossary/Matrix changes are untouched. OCI SSH additional-check session awaits human approval; no policy/TLS bypass.

@Kitkitkittt
Kitkitkittt merged commit 322c115 into main Oct 10, 2026
6 checks passed
@Kitkitkittt

Copy link
Copy Markdown
Owner Author

Follow-up merged as322c115a85581888bb783b75c8d3e52461e8beb4. Synchronized PR CI https://github.com/Kohnnn/vnibb/actions/runs/38021759796 passed all four jobs and nine root steps in520.2s:157 suites/1044 frontend tests,1621 backend passes/four skips. Standards and Spec reviews found zero findings. Main CI https://github.com/Kohnnn/vnibb/actions/runs/38023941212 running; Vercel exact-SHA production rebuild building. OCI access check expired after900s without human approval; current public API still2c576542 with DB/cache connected. No backend service or access-policy change occurred.

This branch was successfully deployed

1 active deployment
Preview — 197ca533 Deployed Oct 10, 2026 by vercel[bot]
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.

1 participant