[timothyanderson096-ocdealcheck]

Summary of Changes

Importing asammdf first can re-enter can through canmatrix/canopen before asammdf is initialized. The optional-import guard then permanently treats an installed dependency as missing.

• Load asammdf when an MF4 reader or writer is constructed.
• Retain the optional-dependency message when asammdf itself is absent; preserve the original error for missing transitive dependencies.
• Resolve writer dependency metadata before opening its output.
• Add fresh-process reader/writer regressions for both native import orders and a deterministic circular import around the real asammdf loader. The deterministic case does not require canopen in upstream test dependencies.
• Add missing-dependency/file-opening controls and a changelog fragment. Frame parsing and serialization are unchanged.

Related Issues / Pull Requests

Closes #2122. Thanks to @nehalkpatel for the reproduction and dependency-chain analysis.

Existing packed-error-frame PR #2115 is separate and is not included or stacked here.

Type of Change

• Bug fix
• New feature
• Documentation update
• Refactoring
• Other

Checklist

• I have followed the contribution guide.
• I have added or updated tests as appropriate.
• I have added or updated documentation as appropriate.
• I have added a news fragment.
• All checks and tests pass (tox).

Validation and limits

Tested on Linux / CPython 3.12.14 with asammdf 8.7.2, canmatrix 1.2 and canopen 2.4.1. Published commit 2c0bb48693d1e96332ef8cb9df74bc9e363e41ed has the exact tested complete tree de8a6291bdd23d56b96df7ffe9baefd09a9a52f3, based on unchanged upstream b4f82abede25ff83376be793a2935c41f81c3869.

• Identical new regressions on the original source: four reader/writer import-order subcases and the transitive-error case fail. Candidate: all three test methods and six subtests pass.
• Focused MF4 checks: 7 passed, 1 existing unsupported-append skip, 6 subtests passed.
• Full local suite: baseline 33 failed, 505 passed, 98 skipped, 31 subtests passed; candidate 28 failed, 506 passed, 98 skipped, 35 subtests passed. The remaining 28 failures are the same IPv4/IPv6 multicast socket configuration failures in both trees. There are no new failing test identities; the full local suite is not green.
• Project-wide Ruff and Black pass. Project mypy targets 3.10–3.14 pass all 65 source files in a lint environment without MF4 extras. Changed-source Pylint and complete-patch whitespace pass.
• An additional mypy diagnostic with the real MF4 dependencies installed retains seven existing-source diagnostics, compared with 16 on the baseline. No type checks or thresholds were weakened.
• Independent coding-agent inspection identified and resolved the initial regression-environment gap; final inspection found no remaining material issue.

No hosted CI, hardware, other OS/interpreter execution, minimum asammdf 6 validation, full tox/docs run or whole-project Pylint result is claimed. Black emits an inferred Python 3.15 target warning under this Python 3.12 runtime; explicit Python 3.10 changed-file checks also pass.

Prepared with OpenAI Codex assistance. Maintainer review remains requested.

Timothy Anderson — Spotted Ridge Engineering / OC Labs · Independent software verification · GDN (Global Decision Network)
timothyanderson096@gmail.com · Engineering evidence

[chatgpt-codex-connector]

Codex Review Summary

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

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-06T15:09:54.382157Z	2c0bb48	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.