pyproject.toml

[build-system]
requires = ["uv_build>=0.9.10"]
build-backend = "uv_build"

[project]
name = "data-foundry"
version = "0.0.5"
description = "A schema and toolkit for curating tabular datasets and benchmarking tasks (the data layer behind TabArena)."
readme = "README.md"
requires-python = ">=3.10"
license = { text = "Apache-2.0" }
authors = [
    { name = "TabArena Maintainers", email = "mail@tabarena.ai" },
]
keywords = [
    "tabular",
    "machine-learning",
    "benchmark",
    "datasets",
    "data-curation",
    "tabarena",
]
classifiers = [
    "Development Status :: 3 - Alpha",
    "Intended Audience :: Science/Research",
    "Intended Audience :: Developers",
    "License :: OSI Approved :: Apache Software License",
    "Operating System :: OS Independent",
    "Programming Language :: Python",
    "Programming Language :: Python :: 3",
    "Programming Language :: Python :: 3.10",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Programming Language :: Python :: 3.13",
    "Topic :: Scientific/Engineering :: Artificial Intelligence",
]
dependencies = [
    "pandas",
    "numpy",
    "pydantic",
    "pyyaml",
    "uuid6",
    "pyarrow",
    "huggingface_hub",
]

[project.scripts]
data-foundry-curation = "data_foundry.curation.cli:main"

[project.urls]
Homepage = "https://github.com/TabArena/data-foundry"
Repository = "https://github.com/TabArena/data-foundry"
Issues = "https://github.com/TabArena/data-foundry/issues"
"BeyondArena Datasets" = "https://huggingface.co/datasets/TabArena/BeyondArena"
"TabArena" = "https://tabarena.ai/"
"Paper" = "https://arxiv.org/abs/2606.30410"

[project.optional-dependencies]
dev = [
    "autogluon",
    "openml",
    "ruff",
    "pyyaml",
    "seaborn",
    "tueplots",
    "tqdm",
    "kaggle",
    "langdetect",
    "xlrd",
    "scipy",
    "polars",
    "fastexcel",
    "openpyxl",
    "python-calamine",
]
tests = [
    "pytest",
    "scikit-learn",
]
curation = [
    # Optional: publish the flat snapshot to a read-only Google Sheet.
    # The editing dashboard itself needs no extra dependencies (stdlib server).
    "gspread",
]

[tool.ruff]
target-version = "py39"
line-length = 120
output-format = "full"
src = ["examples", "tabflow_slurm"]

[tool.ruff.lint]
# Extend what ruff is allowed to fix, even it it may break
# This is okay given we use it all the time and it ensures
# better practices. Would be dangerous if using for first
# time on established project.
extend-safe-fixes = ["ALL"]

# Allow unused variables when underscore-prefixed.
dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"

select = [
    "A",
    # "ANN", # Handled by mypy
    "ARG",
    "B",
    "BLE",
    "COM",
    "C4",
    "D",
    # "DTZ",  # One day I should know how to utilize timezones and dates...
    "E",
    # "EXE", Meh
    "ERA",
    "F",
    "FBT",
    "I",
    # "ISC",  # Favours implicit string concatenation
    "INP",
    # "INT", # I don't understand this one
    "N",
    "NPY",
    "PD",
    "PLC",
    "PLE",
    "PLR",
    "PLW",
    "PIE",
    "PT",
    "PTH",
    # "PYI", # Specific to .pyi files for type stubs
    "Q",
    "PGH004",
    "RET",
    "RUF",
    "C90",
    "S",
    # "SLF",    # Private member accessed (sure, it's python)
    "SIM",
    # "TRY", # Good in principle, would take a lot of work to statisfy
    "T10",
    "T20",
    "TID",
    "TCH",
    "UP",
    "N",
    "W",
    "YTT",
]

ignore = [
    "D104", # Missing docstring in public package
    "D105", # Missing docstring in magic mthod
    "D203", # 1 blank line required before class docstring
    "D205", # 1 blank line between summary and description
    "D401", # First line of docstring should be in imperative mood
    "N806", # Variable X in function should be lowercase
    "E731", # Do not assign a lambda expression, use a def
    "A002", # Shadowing a builtin
    "A003", # Shadowing a builtin
    "S101", # Use of assert detected.
    "W292", # No newline at end of file
    "PLC1901", # "" can be simplified to be falsey
    "TC003", # Move stdlib import into TYPE_CHECKING
    "PLR2004", # Magic numbers, gets in the way a lot
    "PLR0915", # Too many statements
    "N803", # Argument name `X` should be lowercase
    "N802", # Function name should be lowercase
    "COM812", # Trailing comma missing (conflicts with formatter)
    "T201", # Remove prints
    "C408", # use of {} instead of dict
]

exclude = [
    ".bzr",
    ".direnv",
    ".eggs",
    ".git",
    ".hg",
    ".mypy_cache",
    ".nox",
    ".pants.d",
    ".ruff_cache",
    ".svn",
    ".tox",
    ".venv",
    "__pypackages__",
    "_build",
    "buck-out",
    "build",
    "dist",
    "node_modules",
    "venv",
    "docs",
]

[tool.ruff.lint.isort]
no-lines-before = ["future"]
required-imports = ["from __future__ import annotations"]
combine-as-imports = true
extra-standard-library = ["typing_extensions"]
force-wrap-aliases = true

[tool.ruff.lint.pydocstyle]
convention = "google"

[tool.ruff.lint.pylint]
max-args = 10