Repository navigation
P0: integrate realtime FAT + clean evidence on current 1.6.26 main - #159
Conversation
Transplant the validated P0 FAT changes onto the current 1.6.26 main baseline while preserving the timestamp rounding, full-resolution evidence tooltip, and release evidence changes already merged by parallel work.
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 5053bfb951
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".
| var reportCandidates = all | ||
| .Where(point => !IsFastCommissioningPoint(point)) | ||
| .ToList(); |
There was a problem hiding this comment.
When an ordinary Live Monitor project uses a polling interval of 500 ms or less, every Boolean/status/protection point now satisfies IsFastCommissioningPoint and is removed from both static and dynamic report plans, even though no FAT workspace is active. Those users therefore lose configured report acquisition and rely only on sampled MMS reads, which can miss transitions and discard report timestamps; pass an explicit commissioning mode instead of inferring it solely from the polling interval.
Useful? React with 👍 / 👎.
| if (_pollingIntervalBeforeIoFat.HasValue) | ||
| { | ||
| PollingIntervalMs = _pollingIntervalBeforeIoFat.Value; | ||
| _pollingIntervalBeforeIoFat = null; |
There was a problem hiding this comment.
When a FAT-prepared device remains connected after the FAT window closes, assigning the saved PollingIntervalMs only restores the MainWindow setting; StartDeviceMonitorAsync already copied 250 ms into each monitor point, and the current runtime/report plan is not restarted here. Returning to the engineering window therefore leaves that device on high-rate polling with its digital points excluded from reports, so the affected monitors need to be reconfigured or restarted as part of this close path.
Useful? React with 👍 / 👎.
Safe integration of P0 FAT work
This PR supersedes #156 for merge safety. It is built directly on the current
mainbaseline403c451b72fe37ae345c8c154956ab24e379655b, after the 1.6.26 release and its release-evidence synchronization.Only the 11 files changed by the validated P0 FAT work were transplanted onto current main. In particular, this does not replace files changed by parallel progress in #155, #157, #158, or the subsequent 1.6.26 release-evidence sync.
P0 scope:
Historical evidence is not rewritten or deleted.
Merge gate: do not merge until full Windows build/test/package, focused IO FAT tests, and deterministic SV checks all pass on this exact integration head.