pyproject.toml

[build-system]
requires = ["setuptools>=61.0", "wheel"]
build-backend = "setuptools.build_meta"

[project]
name = "ViralScan"
dynamic = ["version"]
description = "A computational framework to detect viral load."
authors = [
    {name="Emma Vonk", email="emma.vonk@hotmail.nl"}
]
license = {file = "LICENSE"}
dependencies = [
    "snakemake>=7.0.0",
    "pyfiglet",
    "scanpy",
    "pandas",
    "plotly",
    "numpy",
    "matplotlib",
    "scipy",
    "seaborn",
    "pyyaml",
    "requests>=2.28",
    "jinja2>=3.0",
    "scikit-learn>=1.0",
    "anndata>=0.9",
]
readme = "README.md"
requires-python = ">=3.9"
classifiers = [
    "Development Status :: 4 - Beta",
    "Intended Audience :: Science/Research",
    "License :: OSI Approved :: MIT License",
    "Operating System :: OS Independent",
    "Programming Language :: Python :: 3",
    "Programming Language :: Python :: 3.9",
    "Programming Language :: Python :: 3.10",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Topic :: Scientific/Engineering :: Bio-Informatics",
]

[project.optional-dependencies]
dev = [
    "pytest>=7",
    "pytest-cov",
    "responses",
    "ruff",
    "mypy",
    "pre-commit",
]
enrichment = [
    "gget>=0.27",
]

[project.scripts]
viralscan = "viralscan.menu:main"

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

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

[tool.setuptools.package-data]
"viralscan" = ["Snakefile", "templates/*.j2"]
"viralscan.data" = ["anellovirus_accessions.tsv"]
"viralscan.scripts" = ["emptydrops.R"]

[project.urls]
"Homepage" = "https://github.com/mdmanurung/ViralScan"
"Bug Tracker" = "https://github.com/mdmanurung/ViralScan/issues"

[tool.pytest.ini_options]
testpaths = ["tests"]
addopts = "-ra -m 'not network and not integration and not research'"
markers = [
    "network: tests that hit the live NCBI API",
    "integration: end-to-end Snakemake smoke tests",
    "research: tests of repo-root analysis scripts (scripts/*), not the installed package",
]

[tool.ruff]
line-length = 100
target-version = "py39"
# Lint the shipped package + its tests only. Repo-root research/analysis/build
# scripts and data dirs are out of scope for the package's lint gate.
extend-exclude = [
    "analysis",
    "covid_viralscan",
    "scripts",
    "benchmark_runs",
    "benchmark_inputs",
    "results",
    "references",
    "docs",
]

[tool.ruff.lint]
# Hardened ruleset (RR4.4): pycodestyle errors + pyflakes + isort (I),
# flake8-bugbear (B), pyupgrade (UP), flake8-simplify (SIM).
select = ["E4", "E7", "E9", "F", "I", "B", "UP", "SIM"]
ignore = [
    "E501",     # line length is enforced by the formatter
    "UP007",    # `X | None` PEP-604 style — cosmetic; `Optional[...]` kept for py39 clarity
    "UP045",    # same, for Optional
    "SIM117",   # nested `with` (readable for stacked patch() in tests)
]

[tool.ruff.lint.per-file-ignores]
# Snakemake script files receive an implicit ``snakemake`` global injected
# by the Snakemake runner; they are not standalone Python modules. They
# also contain a few historical issues that PR 3 will clean up.
"src/viralscan/scripts/analysis.py" = ["F821", "F401"]
"src/viralscan/scripts/createconfig.py" = ["F821"]
"src/viralscan/scripts/detection.py" = ["F821", "F401", "F541", "SIM102"]
"src/viralscan/scripts/multimap.py" = ["F821", "F541"]
"src/viralscan/scripts/umap.py" = ["F821", "F401", "F541", "SIM102"]
"src/viralscan/scripts/host_filter.py" = ["SIM115"]
"src/viralscan/scripts/hostresponse.py" = ["F821"]

[tool.mypy]
python_version = "3.10"
warn_unused_configs = true
# Third-party scientific libs ship no/partial stubs; don't fail on missing imports.
ignore_missing_imports = true

# The scverse/accelerator stack (anndata, scanpy, numba, zarr, …) ships `py.typed`
# and uses PEP 695 (`class Foo[T]`) syntax that mypy rejects while type-checking
# against python_version=3.10 (e.g. anndata>=0.13). Their bodies are not ours to
# check, so skip-follow them — treat as Any without parsing. This keeps the lint
# job robust across upstream releases instead of pinning mypy's target to 3.12
# (which would forfeit 3.10-syntax gating on our own code) or pinning the deps.
[[tool.mypy.overrides]]
module = [
    "anndata.*",
    "scanpy.*",
]
follow_imports = "skip"

[[tool.mypy.overrides]]
module = [
    "viralscan.utils",
    "viralscan.constants",
    "viralscan.menu",
    "viralscan.scripts.ncbi_fetch",
    "viralscan.scripts.build_reference",
]
strict = true

# RR4.3: the 7 Snakemake scripts are type-checked (in non-strict mode) rather than
# blanket-ignored. They import the injected `snakemake` global (untyped), so
# attr-defined/name-defined on that object is tolerated.
[[tool.mypy.overrides]]
module = [
    "viralscan.scripts.analysis",
    "viralscan.scripts.createconfig",
    "viralscan.scripts.detection",
    "viralscan.scripts.multimap",
    "viralscan.scripts.umap",
    "viralscan.scripts.host_filter",
    "viralscan.scripts.hostresponse",
    "viralscan.scripts.cellcalling",
    # research/benchmark tooling, not core pipeline — same relaxed treatment
    "viralscan.reference_strategy",
]
# Type-checked for real errors (arg-type, return-value, assignment, union-attr, …)
# instead of blanket-ignored. Tolerate the injected untyped `snakemake` global and
# annotation-completeness gaps (full annotation is a separate task).
disable_error_code = [
    "name-defined",
    "attr-defined",
    "no-untyped-def",
    "no-untyped-call",
    "type-arg",
]