Cleaned notebooks, added instructions for different scattering geomet…

[JGoodrichBNL]

…ries, and fixed WAXS isointensity ROI bugs

[Copilot]

🟡 Changes recommended

Surface recipe labels can produce incorrect downstream q-z profile counts, and one documentation link is broken.

1 open finding

• Independent q_z centers create spurious z profile bands · New

What changed in this PR

Adds configurable surface-scattering ROIs, fixes WAXS percentile masking, documents geometry setup, and cleans notebook outputs.

Changes:

• Added JSON-based GI-SAXS/GI-WAXS ROI editing and reprojection.
• Restricted WAXS percentiles to valid pixels with regression tests.
• Added geometry guidance and stripped notebook outputs.

File	Description
pyCHX/​_xpcs_pipeline_geometry.py	Adds surface ROI recipe loading and WAXS fix.
pyCHX/​tests/​test_xpcs_pipeline_geometry.py	Tests WAXS and surface ROI behavior.
pipelines/​XPCS_auto.ipynb	Integrates surface ROI recipes.
pipelines/​old_pipelines/​XPCS_SAXS_auto.ipynb	Removes notebook metadata noise.
pipelines/​notes/​beam_center_position_issue.md	Documents beam-center offsets.
pipelines/​masking_and_rois/​Detector_Masks_Add_Pixels.ipynb	Clears outputs and execution state.
pipelines/​masking_and_rois/​Auto_ROI_Mask_Generator.ipynb	Clears outputs and execution state.
pipelines/​GEOMETRY_AND_ROI_SETUP.md	Adds geometry and ROI setup instructions.

🧠 Review effort: Balanced

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

🔵 Needs a closer look

Base ROI validation prevents removing labels that overlap newly masked pixels, and one documentation link is broken.

1 open finding

• Independent q_z centers create spurious z profile bands

Previously missed (1)

In code that hasn't changed since last review

Validate ROI after removing labels

pyCHX/​_xpcs_pipeline_geometry.py:374

The base ROI is validated against the current mask before remove_labels is applied. As a result, a recipe cannot remove a legacy label that now intersects a masked pixel—the loader raises before deleting it—even though the final ROI would be valid. Load the base without mask validation here; the final validate_roi call already checks every retained/new label against mask.

🧠 Review effort: Balanced

[Copilot]

🟡 Changes recommended

Arbitrary surface boxes are converted into synthetic regular bins for downstream GI-SAXS profiles.

1 open finding

• GI-SAXS loses arbitrary ROI bounds by inventing regular bins · New

1 resolved since last review

• Independent q_z centers create spurious z profile bands

🧠 Review effort: Balanced

[Copilot]

🔵 Needs a closer look

Scientific ROI reprojection and profile changes warrant final validation against representative beamline data.

0 open findings

1 resolved since last review

• GI-SAXS loses arbitrary ROI bounds by inventing regular bins

🧠 Review effort: Balanced