- Examples
- Purpose and scope (why this project)
- Serpent axial symmetry rules (more of a rotation in YZ plane)
- Standardize microXS vs microXs
- Guard against hooks that aren't supported by solvers
- Using vectorized / subprocess-managed routines for SFV macroxs reconstruction
- Resolve how hydep.Pin objects are handled by numpy
- Remove most assert statements in favor of actual checks
- Find a way to add regression test file (setup, post-process) into the repository
- Make all tests less flaky. Failures can fill up the Material or Universe register. Some of the writer tests require strict orderings of Materials and Universes.
- Improved parsing of error at output (catch more types of errors)
- Capture ExtDepRunner output on failure. Little broken now
- Support for writing outer materials for lattice stack
- Serpent fluxes, microscopic xs, and macroscopic xs are one group
- Fetching microscopic reaction xs with Serpent requires depletion unless using CoupledSerpentSolver. This class relies on a patched version of Serpent that the author intends to distribute to the user community.
- Multigroup
- OpenMC compatibility?
- Helper function for making pins with automatic subdivision a. la. openmc.model.pin
- Helper for making UO2-like materials
- Weakref for Isotopes
- Share some code between CartesianLattice and LatticeStack
- Use procedurally generated reference files, e.g. with Jinja, for testing Serpent writer
- Add other simple ReducedOrderSolvers that interpolate / extrapolate flux?
- Improve isotope interface. Slapped an lru_cache on the getIsotope function. For a 3x3 pin cell w/ Gad, got ~3800 hits, 17 billion (!!) misses
- Isotopes hashable? They are mutable, in the sense their reactions change. But they are hashed based on their hopefully unique and immutable ZAI tuple