Skip to content

evidence(OMN-14988): OCC companion — check_no_infra_inmemory_import.sh slash-normalization fix - #5132

Closed
jonahgabriel wants to merge 1 commit into
devfrom
jonah/omn-14988-occ-companion
Closed

jonahgabriel wants to merge 1 commit into
devfrom
jonah/omn-14988-occ-companion

Conversation

@jonahgabriel

Copy link
Copy Markdown
Contributor

OMN-14988 — OCC companion (net-new file only)

Companion for the product fix in OmniNode-ai/omnibase_infra (branch
jonah/omn-14988-slash-normalization, commit c9ca7eb1).

Diff scope: one net-new file, contracts/OMN-14988.yaml. No existing file touched.

What the product PR fixes

scripts/validation/check_no_infra_inmemory_import.sh — the no-infra-inmemory-import
pre-commit hook and CI Lint gate (OMN-7077 / OMN-13419) — reported 8 violations on a
clean dev tree
, all 8 of them entries in the script's own ALLOWLIST. Signal-to-noise
was 0/8.

Two platform facts compound:

  1. BSD grep (macOS) reports the search root src/ as src// in every hit, so paths arrive
    as src//omnibase_infra/.... GNU grep on Linux CI does not.
  2. The normalization ${file//\/\//\/} is bash-version-dependent. Under bash >= 4.3 it
    collapses correctly; under bash 3.2 — which is /usr/bin/env bash on stock macOS, and
    therefore the interpreter this hook actually runs under locally
    — the backslash in the
    replacement word is retained literally, producing src\/omnibase_infra/..., which
    matches no ALLOWLIST entry.

Because the failure needs both, Linux CI never saw it and macOS always did.

Evidence recorded in this contract

Item Proof
dod-omn-14988-gate-clean-tree-exit-zero Clean-tree gate run: RC=1 pre-fix → RC=0 post-fix under bash 3.2, on .200 and locally
dod-omn-14988-normalization-regression-red-green 13 failed / 6 passed against the pre-fix script; 19 passed after — same test file both runs
dod-omn-14988-gate-not-neutered A non-allowlisted import with the same src// prefix still exits 1
dod-omn-14988-governed-selection-green Governed selector resolved tests/unit/scripts/; 840 passed / 0 failed on .200
dod-omn-14988-hook-satisfiable-without-bypass Product commit created with the full pre-commit suite running; the OMN-7077 hook passed. No --no-verify, no skip token
dod-omn-14988-shellcheck-clean shellcheck -x clean on the modified script

Ordering

Companion-first. This PR lands before the product PR so the Receipt Gate can resolve
contracts/OMN-14988.yaml.

If OCC autobind opens its own companion for the product PR, autobind wins and this PR is
closed as the duplicate (session precedent).

Merging is Codex's; no --auto set here.

@coderabbitai

coderabbitai Bot commented Jul 27, 2026

Copy link
Copy Markdown

Warning

Review limit reached

You’ve reached a temporary PR review limit under our Fair Usage Limits Policy.

Your recent review volume is higher than typical usage, so adaptive limits are currently applied.

Next review available in: 7 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details
⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 316b4f31-c576-4322-9269-a0d9bd4283fd

📥 Commits

Reviewing files that changed from the base of the PR and between e767d0f and 8974e3c.

📒 Files selected for processing (1)
  • contracts/OMN-14988.yaml
✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Commit unit tests in branch jonah/omn-14988-occ-companion

Comment @coderabbitai help to get the list of available commands.

@jonahgabriel

Copy link
Copy Markdown
Contributor Author

Closing as the duplicate. OCC autobind opened #5133 for the product PR OmniNode-ai/omnibase_infra#2498 three minutes after this one; per standing session precedent autobind wins the companion slot and the hand-authored companion closes.

Note for whoever lands #5133: it is the thin autobind shape — its three dod_evidence items only prove that PR #2498 exists and has files. None of them is falsifiable against the actual fix. The substantive, falsifiable evidence for OMN-14988 (clean-tree exit-0 under bash 3.2, 13-failed→19-passed RED/GREEN on the same test file, anti-neutering check, governed-selection 840/0 on .200, shellcheck clean, and the OMN-7077 hook passing without --no-verify) is recorded in the product PR body and as a dod_evidence comment on OMN-14988.

Those content probes should be re-landed onto contracts/OMN-14988.yaml after #5133 merges — same pattern as #5129 and #5117, which re-landed falsifiable probes displaced by an autobind companion. Filing that as the follow-up rather than racing #5133 with a conflicting edit.

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