Coatings implementation#57
Conversation
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## master #57 +/- ##
==========================================
+ Coverage 73.46% 73.95% +0.49%
==========================================
Files 61 70 +9
Lines 3885 4891 +1006
==========================================
+ Hits 2854 3617 +763
- Misses 1031 1274 +243 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b50da79f50
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".
| λ, | ||
| from_front | ||
| ) where {T, R <: Ray{T}} | ||
| ndir_t, _ = refraction3d(direction(ray), normal, n_incident, n_transmitted) |
There was a problem hiding this comment.
When a splitting coating is hit from a higher-index medium above the critical angle, refraction3d returns the reflected direction with the TIR flag set, but this path discards the flag and still creates a transmitted child with n_transmitted and T_coeff. A coated glass-air boundary with behavior = Splitting() will therefore emit a bogus transmitted branch instead of reflecting all power, corrupting traced beam trees in TIR cases.
Useful? React with 👍 / 👎.
| zrotate3d!(coating, π) | ||
| return RectangularPlateBeamsplitter(substrate, coating) | ||
| M = typeof(coating.model) | ||
| return RectangularPlateBeamsplitter{typeof(width), M}(substrate, coating) |
There was a problem hiding this comment.
If callers pass integer dimensions for width while other dimensions or mesh construction promote to floating point, this explicitly instantiates RectangularPlateBeamsplitter{Int64,...} even though substrate and coating are Prism{Float64,...}/Coating{Float64,...}. For inputs like RectangularPlateBeamsplitter(10, 10, 0.2, n) the constructor will fail at runtime; let T be inferred or use promote_type(typeof(width), typeof(height), typeof(thickness)).
Useful? React with 👍 / 👎.
…em to improve intersection checking.
1 participant
Abstract
This is a new attempt at coatings and is based on #55 as a prerequisite. It will also render #27 obsolete, as polarizing components can be handled using coatings which results in a much more flexible, modular and extensible API.
This should close #30.
Detailed description: Unified Coatings, Face Partitions, and Polarizing Optics
Overview
This PR implements a unified coatings and dynamic boundary physics pipeline in
BeamletOptics.jl. It standardizes boundary interactions (refraction, reflection, splitting) across rays and Gaussian/Astigmatic beamlets, modernizes polarizing components, and optimizes performance to achieve allocation-free tracing.Key Features
CoatingBehavior): Introduced traits (Transmissive,Reflective,Splitting) to cleanly dispatch boundary logic.Uncoated,SimpleARCoating,SimpleHRCoating,SimpleBeamsplitterCoating,JonesCoating, andThinFilmCoating(multi-layer characteristic matrix calculations).:front,:back,:hypotenuse), direction vectors, and spatial predicates to apply localized coatings.power_cutoffthreshold insolve_system!to prevent infinite loops.WaveplateandPolarizingBeamSplitter(plate & cube versions) to sit on top of the coating pipeline. Updated_calculate_global_E0to use correct 3D polarization projection mapping.