pyproject.toml

[build-system]
requires = ["setuptools>=64", "wheel"]
build-backend = "setuptools.build_meta"

[project]
name = "olmix"
description = "A Python project for research and development"
readme = "README.md"
license = {text = "Apache-2.0"}
authors = [
    {name = "Kyle Lo", email = "kylel@allenai.org"}
]
classifiers = [
    "Development Status :: 3 - Alpha",
    "Intended Audience :: Science/Research",
    "License :: OSI Approved :: Apache Software License",
    "Programming Language :: Python :: 3",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Topic :: Scientific/Engineering :: Artificial Intelligence",
]
requires-python = ">=3.11"
dependencies = [
    # Core dependencies
    # Pin to main branch for reproducibility
    "ai2-olmo-core[eval] @ git+https://github.com/allenai/OLMo-core.git@add-inloop-evals",
    "boto3",
    "click",
    "cvxpy",
    "datasets>=3,<4",
    "ecos",
    "gcsfs",
    "lightgbm",
    "matplotlib",
    "numpy",
    "pandas",
    "pyarrow<21",
    "pydantic",
    "pyyaml",
    "s3fs",
    "scikit-learn",
    "scipy",
    "seaborn",
    "statsmodels",
    "torch",
    "tqdm",
    "wandb",
    "yaspin",
]

dynamic = ["version"]

[project.scripts]
olmix = "olmix.cli:cli"
olmix-fit = "olmix.fit.cli:fit"

[project.urls]
Homepage = "https://github.com/allenai/olmix"
Repository = "https://github.com/allenai/olmix"
Issues = "https://github.com/allenai/olmix/issues"

[project.optional-dependencies]
beaker = [
    "beaker-py>=1,<2",
    "GitPython>=3.0,<4.0",
]
dev = [
    # Testing
    "pytest>=7.4.0",
    "pytest-cov>=4.1.0",
    "pytest-asyncio>=0.21.0",
    "pytest-timeout>=2.1.0",
    "flaky>=3.7.0",

    # Code formatting and linting (ruff does both)
    "ruff>=0.1.5",

    # Type checking
    "pyright>=1.1.350",

    # Pre-commit hooks
    "pre-commit>=3.3.3",

    "mypy>=1.0,<1.4",
    "black>=23.1,<24.0",
    "isort>=5.12,<5.13",
]
all = [
    "olmix[dev]",
    "olmix[beaker]",
]

[tool.setuptools.packages.find]
where = ["."]
include = ["olmix*"]

[tool.setuptools.dynamic]
version = {attr = "olmix.version.__version__"}

# Ruff configuration (handles both linting and formatting)
[tool.ruff]
line-length = 120
target-version = "py311"

[tool.ruff.lint]
select = [
    "E",     # pycodestyle errors
    "F",     # pyflakes
    "I",     # isort
    "N",     # pep8-naming
    "W",     # pycodestyle warnings
    "UP",    # pyupgrade
    "B",     # flake8-bugbear
    "C90",   # mccabe complexity
    "RUF",   # ruff specific rules
]
ignore = [
    "E501",  # line too long (handled by formatter)
    "E711",  # comparison to None
    "E712",  # comparison to True/False
    "B007",  # loop control variable not used
    "B008",  # do not perform function calls in argument defaults
    "B023",  # function definition does not bind loop variable
    "B904",  # raise from err
    "B905",  # `zip()` without an explicit `strict=` parameter
    "C901",  # function too complex
    "F811",  # redefinition of unused name
    "N803",  # argument name should be lowercase
    "N806",  # variable in function should be lowercase
    "UP007", # use X | Y for type annotations
    "E402",  # module level import not at top
    "RUF001", # ambiguous unicode character in string
    "RUF003", # ambiguous unicode character in comment
    "RUF012", # mutable class attributes should be annotated with ClassVar
]

[tool.ruff.lint.per-file-ignores]
"__init__.py" = ["F401"]  # Allow unused imports in __init__ files
"tests/*" = ["S101"]       # Allow assert statements in tests

[tool.ruff.lint.mccabe]
max-complexity = 10

[tool.ruff.lint.isort]
known-first-party = ["olmix"]

[tool.ruff.format]
quote-style = "double"
indent-style = "space"
skip-magic-trailing-comma = false

# Pyright configuration
[tool.pyright]
pythonVersion = "3.11"
typeCheckingMode = "basic"
reportMissingImports = false
reportMissingTypeStubs = false
exclude = [
    "**/__pycache__",
    ".venv",
    "build",
    "dist",
    "docs",
]

# Pytest configuration
[tool.pytest.ini_options]
testpaths = ["tests"]
python_files = ["test_*.py", "*_test.py"]
python_classes = ["Test*"]
python_functions = ["test_*"]
addopts = [
    "--strict-markers",
    "--tb=short",
    "--disable-warnings",
    "-v",
]
markers = [
    "unit: Unit tests that don't require external resources",
    "integration: Integration tests that may require external resources",
    "slow: Tests that take a long time to run",
    "gpu: Tests that require GPU",
]
filterwarnings = [
    "ignore::DeprecationWarning",
    "ignore::PendingDeprecationWarning",
]

# Coverage configuration
[tool.coverage.run]
branch = true
source = ["olmix"]
omit = [
    "*/tests/*",
    "*/test_*.py",
    "*/__init__.py",
    "*/conftest.py",
    "*/version.py",
]

[tool.coverage.report]
exclude_lines = [
    "pragma: no cover",
    "def __repr__",
    "if TYPE_CHECKING:",
    "raise AssertionError",
    "raise NotImplementedError",
    "if 0:",
    "if False:",
    "if __name__ == .__main__.:",
    "class .*\\bProtocol\\):",
    "@abstractmethod",
]
precision = 2
show_missing = true
skip_covered = false

[tool.coverage.html]
directory = "htmlcov"