pyproject.toml

[project]
name = "ml-analytics-tools"
version = "0.8.1"
description = "Tools for ML projects and data management"
readme = "README.md"
requires-python = ">=3.11"
# Core dependencies: the data connectors (Snowflake, Redshift, S3, Google
# Sheets, Slack) and their light support libs. The heavy ML/modeling stack
# (mlflow, scikit-learn, catboost, shap, lifelines, plotting) is intentionally
# NOT here — it pins numpy hard and conflicts with host environments such as
# Databricks. Install it via the `[modeling]` extra when you need ModelManager
# or ml_analytics.model_tools.
dependencies = [
    "boto3>=1.37.24",
    "ddtrace>=3.4.1",
    "dotenv>=0.9.9",
    "duckdb>=1.4.1",
    "google-api-python-client>=2.150.0",
    "google-auth>=2.35.0",
    "google-auth-httplib2>=0.2.0",
    "google-auth-oauthlib>=1.2.0",
    "ipykernel>=6.29.5",
    "ipywidgets>=8.1.8",
    "pip>=25.3",
    "polars>=1.30.0",
    "pytest>=8.3.5",
    "pyyaml>=6.0.2",
    "redshift-connector>=2.1.9",
    "ruff>=0.11.4",
    "schedule>=1.2.2",
    "setuptools>=42.0.0",
    "slack-sdk>=3.27.0",
    "snowflake-connector-python[pandas,secure-local-storage]>=4.6.0",
]

[project.optional-dependencies]
# ML/modeling stack used by ModelManager (mlflow) and ml_analytics.model_tools
# (scikit-learn, catboost, shap, lifelines, plotting). Kept optional because
# these pin numpy and conflict with host environments. Version floors are `>=`
# (not `==`) so the resolver can pick numpy-compatible builds.
#   pip install ml-analytics-tools[modeling]
modeling = [
    "mlflow>=3.10.1",
    "mlflow[auth]>=3.10.1",
    "scikit-learn>=1.5.2",
    "catboost>=1.2.8",
    "shap>=0.47.2",
    "lifelines>=0.30.3",
    "seaborn>=0.13.2",
]

# Local development against a remote Databricks cluster/serverless.
# Do NOT install on a Databricks cluster: databricks-connect shadows the
# runtime's built-in PySpark and breaks notebook Spark sessions. On Databricks,
# get_spark() reuses the active session and never needs this.
#   pip install ml-analytics-tools[databricks]
databricks = [
    "databricks-connect>=15.1",
]

[project.scripts]
ml-analytics-auth = "ml_analytics.cli:aws_auth_command"
ml-analytics-sso = "ml_analytics.cli:aws_sso_command"
ml-analytics-tunnel = "ml_analytics.tunnel_manager:main"

[dependency-groups]
dev = [
    "pytest>=8.3.4",
    "mypy>=1.14.1",
    "ruff>=0.9.4",
]

[tool.ruff]
exclude = [
    ".bzr",
    ".direnv",
    ".eggs",
    ".git",
    ".git-rewrite",
    ".hg",
    ".ipynb",
    ".ipynb_checkpoints",
    ".mypy_cache",
    ".nox",
    ".pants.d",
    ".pyenv",
    ".pytest_cache",
    ".pytype",
    ".ruff_cache",
    ".svn",
    ".tox",
    ".venv",
    ".vscode",
    "__pypackages__",
    "_build",
    "buck-out",
    "build",
    "dist",
    "node_modules",
    "site-packages",
    "venv",
]
line-length = 120
indent-width = 4
#target-version = "py313" TODO: parametrize

[tool.ruff.lint]
select = ["E", "F", "I", "UP", "B"]
ignore = [
    "INP001"  # File `docs/conf.py` is part of an implicit namespace package. Add an `__init__.py`.
]
fixable = ["ALL"]
unfixable = []
dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"

[tool.ruff.format]
quote-style = "double"
indent-style = "space"
skip-magic-trailing-comma = false
line-ending = "auto"
docstring-code-format = false
docstring-code-line-length = "dynamic"

[build-system]
requires = ["setuptools>=42", "wheel"]
build-backend = "setuptools.build_meta"

[tool.setuptools]
packages = { find = { include = ["ml_analytics"], exclude = ["img", "img.*"] } }

[tool.mypy]
python_version = "3.11"
show_error_codes = true
show_error_context = true
pretty = true
warn_unused_ignores = true
warn_unused_configs = true
ignore_missing_imports = true
warn_unreachable = true
warn_redundant_casts = true
check_untyped_defs = true
disallow_untyped_calls = true
disallow_incomplete_defs = true
disallow_any_generics = true
disallow_subclassing_any = false
exclude = [
    ".bzr",
    ".direnv",
    ".eggs",
    ".git",
    ".git-rewrite",
    ".hg",
    ".ipynb_checkpoints",
    ".mypy_cache",
    ".nox",
    ".pants.d",
    ".pyenv",
    ".pytest_cache",
    ".pytype",
    ".ruff_cache",
    ".svn",
    ".tox",
    ".venv",
    ".vscode",
    "__pypackages__",
    "_build",
    "buck-out",
    "build",
    "dist",
    "node_modules",
    "site-packages",
    "venv",
]