pyproject.toml

[tool.poetry]
name = "kg-microbe"
version = "0.0.0"
description = "kg-microbe"
authors = ["Harshad Hegde <hhegde@lbl.gov>"]
license = "MIT"
readme = "README.md"

[tool.poetry.dependencies]
python = "^3.10"
click = "*"
importlib-metadata = "^6.8.0"
kghub-downloader = "^0.3.7"
kgx = "^2.4.0"
koza = "*"
biolink-model = "*"
watchdog = "<4.0.0" # ! This is a temporary fix 4.0.0 breaks on GH Actions
oaklib = ">=0.5.25"
requests-cache = "^1.1.0"
llm = "^0.11"
pyobo = ">=0.12.0"
curies = ">=0.8.0"
biopython = "^1.83"
matplotlib = ">=3.7"
toytree = "^3.0"
bacdive = "*"
python-dotenv = "*"
requests-ftp = "^0.3"
pandas = ">=2.0.0"
psutil = "^5.9.0"
duckdb = "^1.5.1"
setuptools = "<81" # ! Keep pkg_resources available for eutils (via oaklib); removed in setuptools 81

[tool.poetry.group.dev.dependencies]
pytest = ">=7.1.2"
tox = ">=3.25.1"
pre-commit = ">=3.3.3"
parameterized = "^0.8.1"
notebook = "^7.0.6"
jupyter-contrib-nbextensions = "^0.7.0"
python-sql = "^1.4.2"
ipython-sql = "^0.5.0"
ruff = ">=0.2.1"
deptry = ">=0.16.0"

[tool.poetry.group.docs]
optional = true

[tool.poetry.group.docs.dependencies]
sphinx = {version = ">=6.1.3"}
sphinx-rtd-theme = {version = ">=1.0.0"}
sphinx-autodoc-typehints = "^1.24.0"
sphinx-click = {version = ">=4.3.0"}
myst-parser = {version = ">=0.18.1"}


[tool.poetry.group.s3.dependencies]
boto3 = "^1.34.37"
aws = "^0.2.5"

[tool.poetry.extras]
docs = [
    "sphinx",
    "sphinx-rtd-theme",
    "sphinx-autodoc-typehints",
    "sphinx-click",
    "myst-parser"
    ]

[tool.poetry.scripts]
kg = 'kg_microbe.run:main'

[tool.poetry-dynamic-versioning]
enable = true
vcs = "git"
style = "pep440"

[tool.black]
line-length = 100
target-version = ["py38", "py39", "py310"]

[tool.ruff]
line-length = 120
target-version = "py310"

[tool.ruff.lint]
extend-ignore = [
    "D211",  # `no-blank-line-before-class`
    "D212",  # `multi-line-summary-first-line`
    "S301",  # `pickle` and modules that wrap it can be unsafe when used to deserialize untrusted data, possible security issue`
    "S605",  # Starting a process with a shell, possible injection detected
    "S202",  # Uses of `tarfile.extractall()` - needed for KGX merge workflow
    ]

# Allow autofix for all enabled rules (when `--fix`) is provided.
fixable = ["ALL"]
unfixable = []

# Select or ignore from https://beta.ruff.rs/docs/rules/
select = [
    "B",  # bugbear
    "D",  # pydocstyle
    "E",  # pycodestyle errors
    "F",  # Pyflakes
    "I",  # isort
    "S",  # flake8-bandit
    "W",  # Warning
]

[tool.ruff.lint.per-file-ignores]
"tests/*" = ["S101"]  # Allow assert in test files
"kg_microbe/query_utils/*" = ["S608"]  # DuckDB queries use database CURIEs, not user input

[tool.ruff.lint.mccabe]
# Unlike Flake8, default to a complexity level of 10.
max-complexity = 10

[tool.ruff.lint.isort]
# Configure known first-party modules for correct import sorting.
# `run` is the repo-root CLI module (run.py); declaring it explicitly keeps
# isort classification deterministic in CI, where filesystem-based first-party
# detection does not pick it up.
known-first-party = ["kg_microbe", "run"]

[tool.pytest.ini_options]
norecursedirs = ["scripts", "notebooks", "docs", ".git"]

[tool.deptry]
extend_exclude = ["notebooks", "docs"]

[tool.codespell]
skip = "*.po,*.ts,.git,pyproject.toml"
count = ""
quiet-level = 3
ignore-words-list = "NiFe,formate"

[build-system]
requires = ["poetry-core>=1.0.0", "poetry-dynamic-versioning"]
build-backend = "poetry_dynamic_versioning.backend"