docs(evidence): record v1.6.40 installed-release Smart Discovery field acceptance - #371
Conversation
…tatic reporting evidence
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: cfff001ea1
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".
|
|
||
| **Status:** Maintainer-accepted stable release; installed-package physical diagnostic recorded on 23 September 2026 (UTC+07:00). | ||
|
|
||
| This record closes the documentation gap between the [physically rejected v1.6.39](../evidence/v1.6.39-physical-rejection.json) and the successfully released v1.6.40 Smart Discovery production route. The maintainer confirms that v1.6.40 was released after successful Smart Discovery physical acceptance. The attached operator diagnostic, summarized here without publishing the raw file, independently identifies the installed v1.6.40 application and pinned engine and shows successful live discovery and static reporting in the recorded session. |
There was a problem hiding this comment.
For any binary rebuilt as version 1.6.40 from a different source revision, this diagnostic still reports the same application identity: scripts/publish-windows-portable.ps1 forces InformationalVersion to the normalized version, and DiagnosticReportBuilder.GetAssemblyVersion reads only that attribute. Since the record also acknowledges that no executable SHA-256 was measured, the diagnostic cannot independently identify commit 1b26dc...; attributing the field result to that exact commit can therefore qualify the wrong build. Record only the observed version, or add an embedded revision/measured artifact hash before making this claim.
AGENTS.md reference: AGENTS.md:L11-L14
Useful? React with 👍 / 👎.
|
|
||
| ## Release and regression boundary | ||
|
|
||
| The [v1.6.39 rejection](../evidence/v1.6.39-physical-rejection.json) documented wrong production compilation, ~31.622 s discovery, and only 23/58 report-backed points. v1.6.40 corrected the production build route via [PR #347](https://github.com/masarray/arsas/pull/347) and was published from the exact source and engine shown above. This installed-package diagnostic confirms the successful Smart Discovery/report-only runtime shape and must **not** be reinterpreted as a pending requalification of the already released baseline. |
There was a problem hiding this comment.
When this record is used to clear the v1.6.39 rejection, the conclusion conflicts with evidence/v1.6.39-physical-rejection.json, whose corrective policy explicitly requires physical retesting of both the portable and installer candidates. This new record covers one installed package and explicitly says the portable executable was not physically exercised, yet this sentence directs readers not to treat requalification as pending. Either attach evidence for both required package tests or narrow the status to a partial installed-package observation.
AGENTS.md reference: AGENTS.md:L215-L227
Useful? React with 👍 / 👎.
Why
The repo has an explicit physical rejection record for v1.6.39 but no comparably direct installed-release field verification record for v1.6.40. That imbalance led an audit to incorrectly treat the already-released Smart Discovery production baseline as still awaiting physical acceptance.
The maintainer confirms the v1.6.40 release followed successful physical Smart Discovery acceptance. A newly supplied installed v1.6.40 diagnostic identifies exact application commit
1b26dc1235b3bac0300bf84c1b5e8fc590934bc5and ARIEC61850648124097621046f5f127ceb1cf853fea54db730, and records 2,395.3 ms Smart Discovery; 204 engine KPI requests, zero duplicates; 58/58 selected static members; 22 Analog + 36 Digital; two active configured RCBs; 58/58 report-backed; zero final unresolved and zero cyclic MMS polling.Changes
evidence/.docs/.Primary unresolved=14vs final runtime unresolved=0; RCB planning availabilityUnknowndespite successful activation; BRCBResvTmsrejection followed by accepted implicit reservation; structured report expansion warnings are not process-value rejection.Boundaries
Verification
Read back all four changed files from the GitHub branch. The evidence JSON parsed and release/engine SHA, 58/58 coverage, zero unresolved/polling, document links, private-IP exclusion and raw-hex exclusion checks passed. Please let repository CI validate the exact PR head before merge.