Create fz-Serpent plugin for Serpent Monte Carlo code

[Copilot]

A Funz plugin for running parametric studies with the Serpent Monte Carlo reactor physics code. Uses serpentTools to parse result files.

Plugin Structure

• .fz/models/Serpent.json - Model configuration with % comments, ${var} syntax, output extraction via serpentTools
• .fz/calculators/Serpent.sh - Calculator script that detects and runs sss2/serpent executable
• .fz/calculators/localhost_Serpent.json - Local calculator configuration

Output Variables

All outputs are formatted as JSON arrays/matrices for easy parsing:

Variable	Description
absKeff	Absorption-based k-effective [value, rel_error] per burnup step
anaKeff	Analog k-effective estimate per burnup step
colKeff	Collision-based k-effective per burnup step
impKeff	Implicit k-effective estimate per burnup step
burnup	Burnup values [MWd/kgU, ...] per burnup step (empty for non-depletion)
burnDays	Burnup time [days] per burnup step (empty for non-depletion)

Example Files

• examples/Serpent/input.inp - UOX fuel pin cell with parameterized enrichment, water density, geometry, and MC settings
• examples/Serpent/input_res.m - Sample Serpent result file for serpentTools parsing validation
• example_usage.ipynb - Jupyter notebook demonstrating plugin usage

CI/Testing

• .github/workflows/ci.yml - CI workflow with JSON validation, shell script checks, fzi/fzc tests, and serpentTools parsing tests
• tests/test_plugin.py - Plugin test suite

Original prompt

Create a repository for Serpent model (see doc at https://serpent.vtt.fi/docs/index.html) to fz. See https://github.com/Funz/fz-Model to get the template of fz "model" plugins (mostly files in .fz/models/... and .fz/calculators/...). Choose a basic test from serpent doc (like UOX_lattice_geometry) and use serpentTools to parse its output (see https://serpent-tools.readthedocs.io/), mainly results prented in https://serpent-tools.readthedocs.io/en/master/examples/ResultsReader.html#results-data

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[yannrichet-asnr]

@copilot in ouptut, also get burnup and burnDays output, that should be arrays or matrix (like all previous results). Check that all these output are formatted as json matrix/array, for latter parsing (use json dump if needed).

[Copilot]

@copilot in ouptut, also get burnup and burnDays output, that should be arrays or matrix (like all previous results). Check that all these output are formatted as json matrix/array, for latter parsing...

Added burnup and burnDays output variables in commit 816b031. All outputs are now formatted as JSON arrays using json.dumps():

• k-effective values: [value, rel_error] per burnup step (2D array for depletion cases)
• burnup: [[MWd/kgU, ...], ...] per burnup step
• burnDays: [[days], ...] per burnup step

Empty arrays [] returned for non-depletion cases.

[yannrichet-asnr]

@copilot Also, the notebook is missing, as well as the CI (like in the template)

[Copilot]

@copilot Also, the notebook is missing, as well as the CI (like in the template)

Added in commit ea0efef:

• example_usage.ipynb - Jupyter notebook demonstrating variable parsing, input compilation, and serpentTools result parsing
• .github/workflows/ci.yml - CI workflow with JSON validation, shell script checks, fzi/fzc tests, and serpentTools parsing tests