Conversation
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 063f05c796
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".
| _smoothedVoltageVectors = SmoothVectors(_smoothedVoltageVectors, voltageVectors, elapsedMilliseconds); | ||
| _smoothedCurrentVectors = SmoothVectors(_smoothedCurrentVectors, currentVectors, elapsedMilliseconds); |
There was a problem hiding this comment.
Commit the exact final analyzer result
When a fast scrub ends, the final request calls ShowPhasors only once and the composition pump then stops, so these eased arrays remain between the previous and requested frame indefinitely; no render callback advances them to the target. The phasor labels/vectors therefore report values from neither frame after the user releases the cursor, and ComtradeHarmonicsWorkstationView.ShowSpectra has the same issue. Bypass easing for the final request or continue animation until committing the exact target.
AGENTS.md reference: AGENTS.md:L103-L106
Useful? React with 👍 / 👎.
|
|
||
| public static double Smooth(double current, double target, double elapsedMilliseconds, double timeConstantMilliseconds) | ||
| { | ||
| if (!double.IsFinite(target)) return current; |
There was a problem hiding this comment.
Reject invalid targets instead of retaining stale values
When a later COMTRADE result contains a non-finite magnitude but has the same channel topology, returning current converts that invalid target into the previous finite value. Consequently, the existing finite-value filtering in PreparePanel cannot remove the vector, and the UI displays a stale magnitude under the new cursor/reference rather than indicating that the new result is invalid; non-finite targets must remain invalid or cause that vector to be dropped.
AGENTS.md reference: AGENTS.md:L60-L66
Useful? React with 👍 / 👎.
Superseded because this PR was accidentally based on
main(1.6.35-era baseline) instead of the field-proventrial/scl-golden-wire-v1636branch. Do not use its artifacts for field testing. The targeted fix has been transplanted onto the correct v1.6.36 field-proven branch without carrying the old baseline.