pyproject.toml

[project]
name = "resolvekit"
version = "0.1.12"
description = "Entity and place resolution system that maps messy place/entity strings and codes to canonical entities"
requires-python = ">=3.12"

license = "MIT"
license-files = ["LICENSE"]
readme = "README.md"
authors = [
    {name = "Jorge Rivera", email = "jorge.rivera@one.org"}
]

classifiers = [
    "Development Status :: 4 - Beta",
    "Intended Audience :: Developers",
    "Intended Audience :: Science/Research",
    "Operating System :: OS Independent",
    "Programming Language :: Python :: 3",
    "Programming Language :: Python :: 3.12",
    "Programming Language :: Python :: 3.13",
    "Programming Language :: Python :: 3.14",
    "Programming Language :: Python :: 3 :: Only",
    "Topic :: Scientific/Engineering :: GIS",
    "Topic :: Scientific/Engineering :: Information Analysis",
    "Topic :: Text Processing :: Linguistic",
    "Typing :: Typed"
]

keywords = [
    "entity-resolution",
    "geocoding",
    "place-names",
    "data-commons",
    "iso-codes",
    "offline",
    "disambiguation",
    "normalization"
]

dependencies = [
    "packaging>=24.0",
    "pooch>=1.8.0,<2.0",
    "pydantic>=2.9,<3.0",
    "rapidfuzz>=3.0.0,<4.0",
    "symspellpy>=6.7.7,<7.0",
]

[project.optional-dependencies]
calibration = [
    "gecko-syndata>=0.6.4",
    "scikit-learn>=1.8.0",
]
data = [
    "babel>=2.18.0",
    "datacommons-client>=2.1.4",
    "pycountry>=23.12.11",
    "pyyaml>=6.0",
]
pandas = [
    "pandas>=2.0",
]
polars = [
    "polars>=1.0",
]
parsing = [
    "ahocorasick_rs>=1.0.3,<2.0",
]
baselines = [
    "spacy>=3.7,<4.0",
]

[project.urls]
Homepage = "https://github.com/jm-rivera/resolvekit"
Repository = "https://github.com/jm-rivera/resolvekit"
Documentation = "https://jm-rivera.github.io/resolvekit/"
Issues = "https://github.com/jm-rivera/resolvekit/issues"

[build-system]
requires = ["uv_build>=0.11.7,<0.12.0"]
build-backend = "uv_build"


[tool.uv]
# gecko-syndata (calibration extra, dev-only) caps numpy<2 conservatively, but
# imports nothing removed in numpy 2. Override the cap so the project resolves
# numpy 2.x, which also has cp313 wheels for Windows.
override-dependencies = ["numpy>=2.1"]

[tool.uv.build-backend]
# Two classes of files are excluded from the sdist and wheel:
#
# 1. Datapack build provenance (changelog/diff/qa/reports) is regenerated on
#    every build and not read at runtime.
# 2. Remote-distribution module assets — the sqlite + symspell.dict for the
#    admin1..5 and cities geo tiers. These ship from the GitHub Release and
#    are fetched on first use by ``resolvekit.core.remote``; the wheel only
#    carries each remote module's ``metadata.json`` so the client knows what
#    to download. Local copies (when produced by a build) are also kept out
#    by .gitignore.
#
# Runtime essentials kept in the wheel: entities.sqlite + symspell.dict for
# bundled tiers, every metadata.json, and *_calibrator.json.
source-exclude = [
    "src/resolvekit/builder/data",
    "src/resolvekit/builder/data/**",
    "src/resolvekit/_data/**/changelog.md",
    "src/resolvekit/_data/**/diff_*.json",
    "src/resolvekit/_data/**/qa_report.json",
    "src/resolvekit/_data/**/reports",
    "src/resolvekit/_data/**/reports/**",
    "src/resolvekit/_data/**/*.sqlite.gz",
    "src/resolvekit/_data/**/*.dict.gz",
    "src/resolvekit/_data/**/*.sqlite-shm",
    "src/resolvekit/_data/**/*.sqlite-wal",
    "src/resolvekit/_data/geo/admin1/entities.sqlite",
    "src/resolvekit/_data/geo/admin1/symspell.dict",
    "src/resolvekit/_data/geo/admin2/entities.sqlite",
    "src/resolvekit/_data/geo/admin2/symspell.dict",
    "src/resolvekit/_data/geo/admin3/entities.sqlite",
    "src/resolvekit/_data/geo/admin3/symspell.dict",
    "src/resolvekit/_data/geo/admin4/entities.sqlite",
    "src/resolvekit/_data/geo/admin4/symspell.dict",
    "src/resolvekit/_data/geo/admin5/entities.sqlite",
    "src/resolvekit/_data/geo/admin5/symspell.dict",
    "src/resolvekit/_data/geo/cities/entities.sqlite",
    "src/resolvekit/_data/geo/cities/symspell.dict",
]
wheel-exclude = [
    "src/resolvekit/builder/data",
    "src/resolvekit/builder/data/**",
    "src/resolvekit/_data/**/changelog.md",
    "src/resolvekit/_data/**/diff_*.json",
    "src/resolvekit/_data/**/qa_report.json",
    "src/resolvekit/_data/**/reports",
    "src/resolvekit/_data/**/reports/**",
    "src/resolvekit/_data/**/*.sqlite.gz",
    "src/resolvekit/_data/**/*.dict.gz",
    "src/resolvekit/_data/**/*.sqlite-shm",
    "src/resolvekit/_data/**/*.sqlite-wal",
    "src/resolvekit/_data/geo/admin1/entities.sqlite",
    "src/resolvekit/_data/geo/admin1/symspell.dict",
    "src/resolvekit/_data/geo/admin2/entities.sqlite",
    "src/resolvekit/_data/geo/admin2/symspell.dict",
    "src/resolvekit/_data/geo/admin3/entities.sqlite",
    "src/resolvekit/_data/geo/admin3/symspell.dict",
    "src/resolvekit/_data/geo/admin4/entities.sqlite",
    "src/resolvekit/_data/geo/admin4/symspell.dict",
    "src/resolvekit/_data/geo/admin5/entities.sqlite",
    "src/resolvekit/_data/geo/admin5/symspell.dict",
    "src/resolvekit/_data/geo/cities/entities.sqlite",
    "src/resolvekit/_data/geo/cities/symspell.dict",
]

[dependency-groups]
dev = [
    "uv-build>=0.11.7",
    "pre-commit>=4.6.0",
    "ruff>=0.15.12",
    "ty>=0.0.32",
    "pandas-stubs>=3.0.0.260204",
    "pandas>=2.0",
    "polars>=1.40.1",
]
test = [
    "pytest>=9.0.3",
    "pytest-benchmark>=5.2.3",
    "pytest-cov>=7.1.0",
    "pytest-xdist>=3.8.0",
]

data = [
    "babel>=2.18.0",
    "datacommons-client>=2.1.4",
    "pycountry>=23.12.11",
    "pyyaml>=6.0",
]

benchmark = [
    "pycountry>=23.12.11",
    "country-converter>=1.2",
    "countryguess>=0.3.0",
    "hdx-python-country>=3.8.0",
    "geonamescache>=2.0.0",
    "rapidfuzz>=3.0.0",
    "psutil>=5.9.0",
    "polars>=1.0.0",
    "datacommons-client>=2.1.4",
]

[tool.ruff]
# Set the maximum line length
line-length = 88

# Target Python 3.12+
target-version = "py312"

# Exclude common directories
exclude = [
    ".git",
    ".venv",
    "venv",
    "__pycache__",
    "build",
    "dist",
    "*.egg-info",
    ".pytest_cache",
    "htmlcov",
]

[tool.ruff.lint]
# Enable recommended rules plus extras
select = [
    "E",     # pycodestyle errors
    "W",     # pycodestyle warnings
    "F",     # pyflakes
    "I",     # isort
    "N",     # pep8-naming
    "UP",    # pyupgrade
    "B",     # flake8-bugbear
    "C4",    # flake8-comprehensions
    "SIM",   # flake8-simplify
    "PL",    # pylint
    "RUF",   # ruff-specific rules
]

# Ignore specific rules that might be too strict
ignore = [
    "E501",    # Line too long (handled by formatter)
    "PLR0913", # Too many arguments
    "PLR2004", # Magic value used in comparison
    "PLC0415", # import not at top-level (intentional: lazy imports for optional deps + circular-import avoidance)
    "PLW0603", # global statement (intentional: module-level singletons for default resolver / config)
    "PLR0915", # too many statements (existing complex functions accepted)
    "PLR0912", # too many branches (existing complex functions accepted)
]

# Allow autofix for all enabled rules
fixable = ["ALL"]
unfixable = []

[tool.ruff.lint.per-file-ignores]
# Allow test files to have more relaxed rules
"tests/**/*.py" = ["E501", "PLR2004", "B017"]

[tool.ruff.format]
# Use double quotes for strings
quote-style = "double"

# Use spaces for indentation
indent-style = "space"

# Like Black, respect magic trailing commas
skip-magic-trailing-comma = false

# Like Black, automatically detect line endings
line-ending = "auto"

[tool.ty.src]
include = [
    "src/resolvekit/__init__.py",
    "src/resolvekit/_convenience.py",
    "src/resolvekit/core",
    "src/resolvekit/packs",
    "src/resolvekit/shared",
    "scripts",
]

[tool.ty.environment]
python-version = "3.12"
root = ["./src", "."]

[tool.ty.terminal]
error-on-warning = true

[tool.pytest.ini_options]
# Test discovery patterns
testpaths = ["tests"]
python_files = ["test_*.py"]
python_classes = ["Test*"]
python_functions = ["test_*"]

# Test execution options
addopts = [
    "-v",
    "--strict-markers",
    "--strict-config",
    "--tb=short",
]

# Warning filters - treat ResourceWarnings as errors to catch connection leaks
filterwarnings = [
    "error::ResourceWarning",
    "ignore:unclosed database.*sqlite3.Connection:ResourceWarning",
]

# Custom markers for organizing tests
markers = [
    "slow: marks tests as slow (deselect with '-m \"not slow\"')",
    "integration: marks tests as integration tests",
    "unit: marks tests as unit tests",
    "matcher: marks tests related to matcher functionality",
    "calibration: marks tests related to confidence calibration",
    "temporal: marks tests related to temporal validation",
    "requires_remote_data: marks tests that need remote data packs not bundled in the wheel (run with --run-remote-data)",
]

# Ignore patterns
norecursedirs = [
    ".git",
    ".venv",
    "venv",
    "__pycache__",
    "*.egg-info",
    ".pytest_cache",
    "htmlcov",
]

[tool.coverage.run]
source = ["src/resolvekit"]
omit = [
    "*/tests/*",
    "*/__pycache__/*",
]

[tool.coverage.report]
precision = 2
exclude_lines = [
    "pragma: no cover",
    "def __repr__",
    "raise AssertionError",
    "raise NotImplementedError",
    "if __name__ == .__main__.:",
    "if TYPE_CHECKING:",
    "@abstractmethod",
]