Sofast diagnostics

[rcbrost]

Added Rendering of Mirror Embedded Surface, Also Bug Fixes

There has been a longstanding need to depict the relationship between a mirror's defining boundary on the (x, y) plane and its corresponding bound on the 3-d embedding optical surface. There also has been a longstanding need to draw slices of both parametric surfaces and measured point mirrors (the latter beyond the scope of this pull request). These new capabilities are also currently needed for a current challenging measurement problem. There also was a need to update and repair key test routines exercising mirror output.

Summary of changes

I extended the mirror rendering code to show the embedded surface of parametric mirrors, and also to show the relationships between defining features on the (x, y) plane and their corresponding features lifted onto the mirror embedding surface. This results in "projected" and "lifted" instances of the geometry.

This also includes implementation of general constant-x and constant-y slices of parametric mirrors, including both their projected and lifted geometries, and connecting lines between them.

These are demonstrated and tested in the test_MirrorOutput.py and example_MirrorOutput.py scripts. Further, the example_MirrorOutput.py scripts now verifies its output against expected output, as it should always have done.

I also extended the example_process_single_facet.py script to utilize this capability, to facilitate study of a new challenging measurement example.

I further began to extend the example_process_single_facet.py script to generate diagnostic output showing the progress of its computation to facilitate understanding and debugging. This work is ongoing, but I wanted to release these improvements instead of waiting until the full SOFAST diagnostic quite is operational.

I also improved some documentation and fixed some bugs encountered, most notably some circular imports which were blocking computation.

Implementation notes

It is possible that my repair to the circular references may have introduced a problem with the SpotAnalysis jupyter notebooks; I did not test these. However, the reason I conjecture that this might have occurred is due to a very large and problematic import dependence which caused the observed circular import problem. If problems with the SpotAnalysis jupyter notebooks are observed, then the SpotAnalysis code should be refactored to prevent the introduction of the multiple circular imports. (One specific example I encountered was that when a RegionXY was imported, it correctly imported a LoopXY class which in turn imported the View3d class. This is appropriate, since a RegionXY contains LoopXY objects, and both classes must be able to draw themselves. However, the problem occurred when the import of the View3d class forced an import of the entire rendering package, and then the entire computer vision package, and then large chunks of the SpotAnaylsis system. One of the SpotAnalysis files ultimately needed a RegionXY object, and this caused the circular import loop. This expansive growth of imports is excessive and should be prevented by refactoring.

Submission checklist

• Target branch is develop, not main
• Existing tests are updated or new tests were added
• opencsp/test/test_DocStringsExist.py are verified to include this change or have been updated accordingly
• .rst file(s) under doc/ are verified to include this change or have been updated accordingly

Additional information

Work on SOFAST diagnostics will continue.

[jehsharp]

Hey Randy, in file common/lib/csp/MirrorAbstract.py, on line 842, this string:

If "project_..." is true

is causing the Sphinx doc check to fail, it does not like that syntax. Locally I changed it to something like this:

If any project slice variable is true (ie, project_origin_slice_y)

And that syntax works. Would you mind making that small change and pushing to the PR?

Thanks so much!

[rcbrost]

Hi Jeff, Done! Please let me know if you need anything else. Thanks,

…

-Randy From: Jeff Sharpe ***@***.***> Sent: Monday, February 16, 2026 1:32 PM To: sandialabs/OpenCSP ***@***.***> Cc: Brost, Randy ***@***.***>; Author ***@***.***> Subject: [EXTERNAL] Re: [sandialabs/OpenCSP] Sofast diagnostics (PR #328) @jehsharp requested changes on this pull request. Hey Randy, in file common/lib/csp/MirrorAbstract.py, on line 842, this string: If "project_..." is true is causing the Sphinx doc check to fail, it does not like that syntax. Locally I changed it to something like this: If any project slice variable is true (ie, project_origin_slice_y) And that syntax works. Would you mind making that small change and pushing to the PR? Thanks so much! - Reply to this email directly, view it on GitHub<#328 (review)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BCESO332B65UUR6DVNQRMC34MISKRAVCNFSM6AAAAACVG6FWG6VHI2DSMVQWIX3LMV43YUDVNRWFEZLROVSXG5CSMV3GSZLXHMZTQMJQGQZDEMRXGA>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[jehsharp]

This looks good to me!