pyproject.toml

[project]
name = "kreuzberg"
version = "3.1.4"
description = "A text extraction library supporting PDFs, images, office documents and more"
readme = "README.md"
keywords = [
  "document-processing",
  "image-to-text",
  "ocr",
  "pandoc",
  "pdf-extraction",
  "rag",
  "table-extraction",
  "tesseract",
  "text-extraction",
  "text-processing",
]
license = { text = "MIT" }
authors = [ { name = "Na'aman Hirschfeld", email = "nhirschfed@gmail.com" } ]
requires-python = ">=3.9"
classifiers = [
  "Development Status :: 4 - Beta",
  "Intended Audience :: Developers",
  "License :: OSI Approved :: MIT License",
  "Operating System :: OS Independent",
  "Programming Language :: Python :: 3 :: Only",
  "Programming Language :: Python :: 3.9",
  "Programming Language :: Python :: 3.10",
  "Programming Language :: Python :: 3.11",
  "Programming Language :: Python :: 3.12",
  "Programming Language :: Python :: 3.13",
  "Topic :: Scientific/Engineering :: Artificial Intelligence",
  "Topic :: Software Development :: Libraries :: Python Modules",
  "Topic :: Text Processing :: General",
  "Topic :: Utilities",
  "Typing :: Typed",
]

dependencies = [
  "anyio>=4.9.0",
  "charset-normalizer>=3.4.1",
  "exceptiongroup>=1.2.2; python_version<'3.11'",
  "html-to-markdown>=1.3.2",
  "playa-pdf>=0.4.1",
  "pypdfium2==4.30.0",                                # pinned due to bug in 4.30.1, until v5 is stable
  "python-calamine>=0.3.2",
  "python-pptx>=1.0.2",
  "typing-extensions>=4.13.2; python_version<'3.12'",
]

optional-dependencies.all = [
  # easyocr
  "easyocr>=1.7.2",
  # gmft
  "gmft>=0.4.1",
  # paddle
  "paddleocr>=2.10.0",
  "paddlepaddle>=3.0.0",
  # chunking
  "semantic-text-splitter>=0.25.1",
  "setuptools>=79.0.1",
]
optional-dependencies.chunking = [
  "semantic-text-splitter>=0.25.1",
]
optional-dependencies.easyocr = [
  "easyocr>=1.7.2",
]
optional-dependencies.gmft = [
  "gmft>=0.4.1",
]
optional-dependencies.paddleocr = [
  "paddleocr>=2.10.0",
  "paddlepaddle>=3.0.0",
  "setuptools>=79.0.1",
]
urls.homepage = "https://github.com/Goldziher/kreuzberg"

[dependency-groups]
dev = [
  "covdefaults>=2.3.0",
  "mypy>=1.15.0",
  "pre-commit>=4.2.0",
  "pytest>=8.3.5",
  "pytest-cov>=6.1.1",
  "pytest-mock>=3.14.0",
  "pytest-timeout>=2.3.1",
  "ruff>=0.11.7",
  "trio>=0.30.0",
  "uv-bump",
]
doc = [
  "mkdocs>=1.6.1",
  "mkdocs-git-revision-date-localized-plugin>=1.4.5",
  "mkdocs-material[imaging]>=9.6.9",
  "mkdocs-minify-plugin>=0.8.0",
  "mkdocstrings[python]>=0.29.0",
]

[tool.setuptools.package-data]
kreuzberg = [ "py.typed" ]

[tool.ruff]
target-version = "py39"
line-length = 120
src = [ "kreuzberg", "tests" ]
format.docstring-code-line-length = 120
format.docstring-code-format = true
lint.select = [ "ALL" ]
lint.ignore = [
  "ANN401", # Dynamically typed ANY for kwargs
  "COM812", # Conflicts with formatter
  "D100",   # Missing docstring in public module
  "D104",   # Missing docstring in public package
  "D107",   # Missing docstring in __init__
  "D205",   # 1 blank line required between summary line and description
  "E501",   # Line too long, handled by ruff format
  "EM",     # Exception messages,
  "FBT",    # Boolean-typed positional argument in function definition
  "FIX",    # We allow todo and fixme comments
  "ISC001", # Conflicts with formatter
  "TD",     # We allow todo and fixme comments
  "TRY",    # Try except block, rules are too strict
]
lint.per-file-ignores."tests/**/*.*" = [
  "ARG001",
  "D",
  "N815",
  "PD",
  "PGH003",
  "PLR0915",
  "PLR2004",
  "PT006",
  "PT007",
  "PT013",
  "RUF012",
  "S",
  "SLF001",
]
lint.isort.known-first-party = [ "kreuzberg", "tests" ]
lint.mccabe.max-complexity = 15
lint.pydocstyle.convention = "google"
lint.pylint.max-args = 10
lint.pylint.max-branches = 15
lint.pylint.max-returns = 10

[tool.pyproject-fmt]
keep_full_version = true
max_supported_python = "3.13"

[tool.pytest.ini_options]
filterwarnings = [
  "ignore:Exception ignored in:pytest.PytestUnraisableExceptionWarning",
  "ignore:pkg_resources is deprecated as an API:DeprecationWarning",
  "ignore:ast.Num is deprecated and will be removed in Python 3.14:DeprecationWarning",
  "ignore:datetime.datetime.utcfromtimestamp() is deprecated:DeprecationWarning",
  "ignore:Deprecated call to `pkg_resources.declare_namespace('google')`:DeprecationWarning",
]

[tool.coverage.run]
omit = [ "tests/*", "scripts/*" ]
plugins = [ "covdefaults" ]
source = [ "kreuzberg" ]

[tool.coverage.report]
exclude_lines = [ 'if TYPE_CHECKING:', "except ImportError:" ]
fail_under = 95

[tool.mypy]
packages = [ "kreuzberg", "tests" ]
python_version = "3.9"
disable_error_code = 'import-untyped'
implicit_reexport = false
show_error_codes = true
strict = true

[tool.uv.sources]
uv-bump = { git = "https://github.com/Goldziher/uv-bump" }