.gitignore

# Ignore data files
data/*
# Ignore compiled files across the repo
**/__pycache__/
# Ignore pycharm project files
/.idea/
/.idea_old/
# Ignore .ipynb files
.ipynb

# PyPi
dist/*
caer.egg-info/*
pythonenv3.8/*
build/*
*.pyc
.Python
*.egg-info/
develop-eggs/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
.installed.cfg
*.egg
MANIFEST

.DS_Store
.env

# Virtual environment
.venv/

# Tooling caches and coverage artifacts
.pytest_cache/
.ruff_cache/
.coverage
.coverage.*
coverage.xml
htmlcov/

# Particle tracking output
*.npy
debug.log

# Test artifacts
test_results/

# Run outputs.
#
# These patterns are deliberately scoped to the run-output directories at the
# repository root instead of being repo-wide globs. Repo-wide '*.png', '*.csv',
# '*.dat', '*.h5' and friends used to swallow every figure, benchmark result and
# example dataset a contributor added anywhere in the tree, which is why the
# tracked benchmark results under benchmarks/results/ had to be force-added.
# Put generated artifacts in one of these directories and they stay untracked;
# anything committed on purpose elsewhere stays visible to git.
/output/
/outputs/
/results/
/figures/

# The benchmark harness regenerates a multi-hundred-megabyte synthetic PLOT3D
# grid. benchmarks/.gitignore already covers it; repeated here so the file
# cannot become committable if that file is ever reorganized.
benchmarks/results/synthetic_grid.x

# Videos / animations (large, and never a source artifact)
*.mp4
*.avi
*.mov
*.webm
*.mkv

# Cached particle tracks for the paper figure (regenerate with paper/make_figure.py)
paper/.tracks.npz