[hyperrealist]

/binned also returns the RGB diodes, as DiodesRed/DiodesGreen/DiodesBlue on the camera's bins, so the UI can switch between camera and diodes (DiamondLightSource/atlas#235). It's additive: RedTotal/GreenTotal/BlueTotal are unchanged, and a run without diodes answers null for them.

The diodes are read from primary/spectroscopy_diodes-<colour> or primary/<colour> (fly scans, the PandA's per-frame means), else from the step scan's event table primary/internal/spectroscopy_diodes-<colour>. They don't take part in reconciling lengths, so a lagging PandA never holds the camera images back.

Testing

• tox, with new tests for fly, step (a real event table), lagging diodes, the scan grid and runs without diodes.
• Copies of b01-1 runs cb4c7fc4 (fly) and dbc25a90 (step) in a local server: the diode images match numpy's binning on the same edges. A recent run without diodes answers null.

🤖 Generated with Claude Code

https://claude.ai/code/session_01BsMDFhupHcBFbVJPK8QAvR

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.49%. Comparing base (a2093f6) to head (d0d252d).

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #24      +/-   ##
==========================================
+ Coverage   95.35%   95.49%   +0.13%     
==========================================
  Files           5        5              
  Lines         775      799      +24     
==========================================
+ Hits          739      763      +24     
  Misses         36       36              

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.