pyproject.toml

[build-system]
requires = ["setuptools>=68", "wheel"]
build-backend = "setuptools.build_meta"

[project]
name = "dataset-audit-kit"
version = "0.3.6"
description = "A lightweight dataset validation toolkit for schema, missingness, and drift checks."
readme = "README.md"
requires-python = ">=3.10"
license = { text = "MIT" }
authors = [
  { name = "Muhtasim Munif Fahim" }
]
keywords = ["data quality", "validation", "ml", "dataset", "drift"]
dependencies = [
  "pandas>=2.2.0",
  "pyarrow>=17.0.0",
  # The `hist` subcommand bins values with numpy directly, so it is a
  # first-order dependency rather than something inherited from pandas.
  "numpy>=1.26",
  # YAML serialization for --yaml-out and AuditReport.to_yaml.
  "PyYAML>=6.0",
]
classifiers = [
  "Development Status :: 3 - Alpha",
  "Intended Audience :: Developers",
  "License :: OSI Approved :: MIT License",
  "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 :: Information Analysis",
]

[project.optional-dependencies]
# Object-store backends. pandas delegates s3:// and gs:// URLs to fsspec, which
# needs the matching filesystem package installed.
remote = [
  "fsspec>=2024.6.0",
  "s3fs>=2024.6.0",
  "gcsfs>=2024.6.0",
]
dev = [
  "pytest>=8.0.0",
  "scikit-learn>=1.5.0",
  "nbformat>=5.10.0",
  "openpyxl>=3.1.0",
]

[project.scripts]
dataset-audit-kit = "dataset_audit_kit.cli:main"

[project.urls]
Homepage = "https://github.com/Muhtasim-Munif-Fahim/dataset-audit-kit"
Repository = "https://github.com/Muhtasim-Munif-Fahim/dataset-audit-kit"
Issues = "https://github.com/Muhtasim-Munif-Fahim/dataset-audit-kit/issues"

[tool.setuptools]
package-dir = { "" = "src" }

[tool.setuptools.packages.find]
where = ["src"]

[tool.pytest.ini_options]
testpaths = ["tests"]
addopts = "-q"