pyproject.toml

[build-system]
requires = ["hatchling>=1.27", "hatch-vcs"]
build-backend = "hatchling.build"

[project]
# Storing project metadata in pyproject.toml https://peps.python.org/pep-0621/
name = "voc4cat"
description = "SKOS vocabulary management tool by NFDI4Cat."
authors = [
  # Authors sorted by number of commits as of 2023-07-04
  {name = "David Linke", email = "david.linke@catalysis.de"},
  {name = "Peter Philips", email = "peter.philips@surroundaustralia.com"},
  {name = "Nicholas Car", email = "nick@kurrawong.net"},
  {name = "Jamie Feiss", email = "jamie.feiss@surroundaustralia.com"},
]
maintainers = [
  {name = "David Linke", email = "david.linke@catalysis.de"},
]
license = "BSD-3-Clause"
license-files = [ "LICENSE" ]
readme = "README.md"
requires-python = ">=3.10"

keywords = ["SKOS", "vocabulary", "spreadsheet", "xlsx", "linked data", "rdf"]

classifiers = [
  "Development Status :: 5 - Production/Stable",
  "Programming Language :: Python",
  "Environment :: Console",
  "Intended Audience :: Science/Research",
  "Operating System :: OS Independent",
  "Programming Language :: Python :: 3.10",
  "Programming Language :: Python :: 3.11",
  "Programming Language :: Python :: 3.12",
  "Programming Language :: Python :: 3.13",
  "Programming Language :: Python :: 3.14",
]

dependencies = [
  "base32-crockford",
  "colorama",
  "curies",
  "jinja2",
  "openpyxl >= 3.1.5",
  "pillow",  # for supporting images in xlsx files
  "pydantic >= 2.10",
  "pyshacl",
  "rdflib",
  "tomli>=1.1.0; python_version < '3.11'",
  "isodate >=0.7.2, <1.0.0; python_version < '3.11'",
]

dynamic = ["version"]

[project.urls]
Homepage = "https://github.com/nfdi4cat/voc4cat-tool/"
Documentation = "https://github.com/nfdi4cat/voc4cat-tool/"
Changelog = "https://github.com/nfdi4cat/voc4cat-tool/blob/main/CHANGELOG.md"

[project.optional-dependencies]
tests = [
  "pytest",
  "pytest-subprocess",
  "coverage",
]
lint = [
  "ruff",
  "mypy",
]
docs = [
  "sphinx",
  "myst-parser",
  "furo",
  "sphinx-sitemap",
]
dev = [
    # Recursively including the project's own optional dependencies requires pip>=21.2
    "voc4cat[tests,lint,docs]",
]
assistant = [
  "click",
  "sentence-transformers",
  "torch",
  "levenshtein",
]

[project.scripts]
voc4cat = "voc4cat.cli:run_cli_app"
voc4cat-merge = "voc4cat.merge_vocab:main_cli"
voc-assistant = "voc4cat.assistant:cli"

[tool.hatch.metadata]
# Hatch disallows direct references for dependencies by default.
# We need to enable them to allow installing NFDI4Cat's vocexcel from github.
allow-direct-references = true

# Build targets for hatch are defined as sections within tool.hatch.build.targets:
[tool.hatch.build.targets.sdist]
include = [
  "/src",
  "/templates",
]
[tool.hatch.build.targets.wheel]
packages = ["src/voc4cat"]

# integrates git-tag-based versions into hatch, https://github.com/ofek/hatch-vcs
[tool.hatch.version]
source = "vcs"

[tool.hatch.build.hooks.vcs]
version-file = "src/voc4cat/_version.py"

[tool.pytest.ini_options]
# pytest configuration:
# https://docs.pytest.org/en/stable/customize.html

# Sets directories to be searched for tests. Useful when all project tests are
# in a known location to speed up test collection and to avoid picking up
# undesired tests by accident. (related: issue #95)
testpaths = [
  "tests",
 ]

# Directories that are not visited by pytest collector:
norecursedirs = "dist build .tox .git .cache __pycache__ .venv"

# Execute doctests in classes, functions, and test modules (pytest)
# --doctest-modules
addopts = [
    "-ra",
    "--strict-config",   # Raise an error instead of a warning for pytest config issues
    "--strict-markers",  # Treat typos in function markers as an error
    "--tb=short",
    "--durations=10",           # Show 10 slowest tests
    "--durations-min=1.0",      # Only show tests >1s
]

# Filter warnings from third-party dependencies (show each type only once)
# Deprecations added in rdflib 7.5.0 get triggered pyshacl 0.30.1
filterwarnings = [
    "once:Dataset.default_context is deprecated:DeprecationWarning:pyshacl",
    "once:Dataset.default_context is deprecated:DeprecationWarning:rdflib",
    "once:Dataset.contexts is deprecated:DeprecationWarning:rdflib",
    "once:Dataset.identifier is deprecated:DeprecationWarning:rdflib",
]


[tool.coverage]
[tool.coverage.run]
# https://coverage.readthedocs.io/en/latest/config.html
parallel = true
branch = true
# dynamic_context = "test_function"
source = ["voc4cat"]
source_dirs = ["tests"]
omit = [
    "**/voc4cat/_version.py",
    "**/voc4cat/assistant.py",  # no tests yet
]

[tool.coverage.paths]
# Specify where coverage combine should look for source files.
source = [
    "src",
    "**/site-packages", # for not using tox
    # ".tox/**/site-packages",
]

[tool.coverage.report]
# Show in report which lines are not covered
show_missing = false
# Any line of the source code that matches one of these regexes is excluded
skip_covered = true
# from being reported as missing.
exclude_lines = [
    # Have to re-enable the standard pragma
    "pragma: no cover",
    # Don't complain if tests don't hit defensive assertion code:
    "raise AssertionError",
    "raise NotImplementedError",
    "return NotImplemented",
    "if __name__ == .__main__.:",
]

[tool.coverage.html]
# Directory where to write the HTML report files.
directory = ".htmlcov"
title = "voc4cat coverage report"

[tool.ruff]
# https://beta.ruff.rs/docs/configuration/

exclude = [
    "__pycache__",
    "*.egg",
    ".*",
    "_version.py",
]

# Assume Python 3.10 syntax and semantics for linting.
target-version = "py310"

# Same as Black.
line-length = 88

[tool.ruff.lint]

ignore = [
    "B905",  # zip() without an explicit strict= parameter set. (requires python >= 3.10)
    "E501",  # line too long
]

# Avoid trying to fix these violations
unfixable = [
    "B",  # flake8-bugbear
]
# Rule selection https://beta.ruff.rs/docs/rules/
select = [
    # sort order as in ruff docs
    "F",   # Pyflakes
    "E",   # pycodestyle (errors)
    "W",   # pycodestyle (warnings)
    "C90", # mccabe
    "I",   # isort
    "N",   # pep8-naming
    # "D",   # pydocstyle
    "UP",  # pyupgrade
    "YTT", # flake8-2020
    # flake8-annotations (ANN)
    # flake8-async (ASYNC)
    "S",   # flake8-bandit
    "B",   # flake8-bugbear
    "A",   # flake8-builtins
    "C4",  # flake8-comprehensions
    "DTZ", # flake8-datetimez
    "T10", # flake8-debugger
    # "DJ",  # flake8-django
    "EM",  # flake8-errmsg
    "ISC", # flake8-implicit-str-concat
    "G",   # flake8-logging-format
    "PIE", # flake8-pie
    "PT",  # flake8-pytest-style
    "Q",   # flake8-quotes
    "RET", # flake8-return
    "SIM", # flake8-simplify
    # "ARG", # flake8-unused-arguments (ARG)
    # "PTH", # flake8-use-pathlib (PTH)
    # "ERA", # eradicate (ERA) - commented out code
    # pandas-vet (PD)
    "PGH", # pygrep-hooks
    "PL",  # whole Pylint (Convention, Error, Refactor, Warning)
    "TRY", # tryceratops (TRY)
    # NumPy-specific rules (NPY)
    "RUF",  # Ruff-specific rules
]

[tool.ruff.lint.per-file-ignores]
"tests/*.py" = [
  "PLR2004", # Allow numbers instead of named constants in tests
  "S101",   # assert in tests is OK
]
"example/*.py" = [
  "PLR0915", # allow many statements in functions in example scripts
  "S101",   # assert in example scripts is OK
]

[tool.ruff.lint.mccabe]
# Flake8-mccabe uses a default level of 7, ruff of 10.
max-complexity = 10

[tool.ruff.lint.pep8-naming]
# Allow Pydantic's `@validator` decorator to trigger class method treatment.
classmethod-decorators = [
  "classmethod",
]

[tool.mypy]
# https://mypy.readthedocs.io/en/latest/config_file.html#using-a-pyproject-toml-file
# Suppress all missing import errors for all untyped libraries
ignore_missing_imports = true