pyproject.toml

[project]
name = "docling-core"
version = "2.74.1"  # DO NOT EDIT, updated automatically
description = "A python library to define and validate data types in Docling."
license = "MIT"
license-files = ["LICENSE"]
keywords = ["docling", "discovery", "etl", "information retrieval", "analytics", "database", "database schema", "schema", "JSON"]
readme = "README.md"
authors = [
    { name = "Cesar Berrospi Ramis", email = "ceb@zurich.ibm.com" },
    { name = "Panos Vagenas", email = "pva@zurich.ibm.com" },
    { name = "Michele Dolfi", email = "dol@zurich.ibm.com" },
    { name = "Christoph Auer", email = "cau@zurich.ibm.com" },
    { name = "Peter Staar", email = "taa@zurich.ibm.com" },
]
maintainers = [
    { name = "Panos Vagenas", email = "pva@zurich.ibm.com" },
    { name = "Michele Dolfi", email = "dol@zurich.ibm.com" },
    { name = "Christoph Auer", email = "cau@zurich.ibm.com" },
    { name = "Peter Staar", email = "taa@zurich.ibm.com" },
    { name = "Cesar Berrospi Ramis", email = "ceb@zurich.ibm.com" },
]
classifiers = [
    "Development Status :: 5 - Production/Stable",
    "Intended Audience :: Developers",
    "Intended Audience :: Science/Research",
    "Natural Language :: English",
    "Operating System :: OS Independent",
    "Topic :: Database",
    "Topic :: Scientific/Engineering :: Information Analysis",
    "Topic :: Software Development :: Libraries :: Python Modules",
    "Typing :: Typed",
    "Programming Language :: Python :: 3",
    "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",
]
requires-python = '>=3.10,<4.0'
dependencies = [
    'jsonschema (>=4.16.0,<5.0.0)',
    'pydantic (>=2.6.0,<3.0.0,!=2.10.0,!=2.10.1,!=2.10.2)',
    'jsonref (>=1.1.0,<2.0.0)',
    'tabulate (>=0.9.0,<0.11.0)',
    'pandas (>=2.1.4,<4.0.0)',
    'pillow (>=10.0.0,<13.0.0)',
    'pyyaml (>=5.1,<7.0.0)',
    'typing-extensions (>=4.12.2,<5.0.0)',
    'typer (>=0.12.5,<0.25.0)',
    'latex2mathml (>=3.77.0,<4.0.0)',
    "defusedxml (>=0.7.1, <0.8.0)",
    "pydantic-settings>=2.14.0",
]

[project.urls]
homepage = "https://github.com/docling-project"
repository = "https://github.com/docling-project/docling-core"
issues = "https://github.com/docling-project/docling-core/issues"
changelog = "https://github.com/docling-project/docling-core/blob/main/CHANGELOG.md"

[project.scripts]
# validate = "docling_core.utils.validate:main"  # use python -m docling_core.utils.validate
# generate_jsonschema = "docling_core.utils.generate_jsonschema:main"  # use python -m docling_core.utils.generate_jsonschema
# generate_docs = "docling_core.utils.generate_docs:main"  # use python -m docling_core.utils.generate_docs
docling-view = "docling_core.cli.view:app"

[project.optional-dependencies]
chunking = [
    # common:
    'semchunk (>=2.2.0,<4.0.0)',
    'tree-sitter (>=0.25.0,<0.27.0)',
    'tree-sitter-python >=0.23.6',
    'tree-sitter-c >=0.23.4',
    'tree-sitter-javascript >=0.23.1',
    'tree-sitter-typescript >=0.23.2',

    # specific:
    'transformers (>=4.34.0,<6.0.0)',
]
chunking-openai = [
    # common:
    'semchunk (>=2.2.0,<4.0.0)',
    'tree-sitter (>=0.25.0,<0.27.0)',
    'tree-sitter-python >=0.23.6',
    'tree-sitter-c >=0.23.4',
    'tree-sitter-javascript >=0.23.1',
    'tree-sitter-typescript >=0.23.2',

    # specific:
    'tiktoken (>=0.9.0,<0.13.0)',
]
examples = [
    "datasets>=4.0.0",
    "matplotlib>=3.7.0",
    "openpyxl>=3.1.5",
]

[dependency-groups]
dev = [
    "pre-commit~=3.7",
    "mypy~=1.6",
    "black~=26.3.1",
    "isort~=5.10",
    "autoflake~=2.0",
    "flake8~=7.1",
    "pycodestyle~=2.10",
    "flake8-docstrings~=1.6",
    "pep8-naming~=0.13",
    "jsondiff~=2.0",
    "types-setuptools~=70.3",
    "pandas-stubs~=2.1",
    "ipykernel~=6.29",
    "coverage~=7.6",
    "pytest~=8.3",
    "pytest-cov>=6.1.1",
    "python-semantic-release~=7.32",
    'tree-sitter-java-orchard (>=0.3.0,<1.0.0); python_version >= "3.10"',
    "ruff>=0.14.11",
    "types-defusedxml (>=0.7.0.20250822, <0.8.0)",
]
constraints = [
    'pandas (>=2.1.4,<3.0.0); python_version < "3.11"',
    'pandas (>=2.1.4,<4.0.0); python_version >= "3.11"',
]

[tool.uv]
package = true

[tool.setuptools.packages.find]
where = [".", "docling_core/resources/schemas"]
include = ["docling_core*"]
namespaces = true

[tool.setuptools.package-data]
"*" = ["*.json"]

[tool.ruff]
target-version = "py310"
line-length = 120
respect-gitignore = true
exclude = [
    "test/data/**"
]

[tool.ruff.format]
skip-magic-trailing-comma = false

[tool.ruff.lint]
select = [
  "C",     # flake8-comprehensions
  "C9",    # mccabe
  "E",     # pycodestyle errors (default)
  "F",     # pyflakes (default)
  "I",     # isort
  "PD",    # pandas-vet
  "PIE",   # pie
  "Q",     # flake8-quotes
  "RUF",   # Enable all ruff-specific checks
  "S307",  # eval
  "W",     # pycodestyle warnings
  "ASYNC", # async
  "UP",    # pyupgrade
]

ignore = [
  "C408",   # Unnecessary `dict()` call (rewrite as a literal)
  "E501",   # Line too long, handled by ruff formatter
  "E741",   # Ambiguous variable name: `l`
  "D107",   # "Missing docstring in __init__",
  "F401",   # imported but unused; consider using `importlib.util.find_spec` to test for "
  "PL",     # Pylint
  "RUF005", # Consider ... instead of concatenation
  "RUF012", # Mutable Class Attributes
  "UP007",  # Option and Union
  "UP045",  # Use `X | None` for type annotations
]

[tool.ruff.lint.pep8-naming]
classmethod-decorators = [
  "classmethod",
  "validator",
  "pydantic.validator",
]

[tool.ruff.lint.pydocstyle]
convention = "google"

[tool.ruff.lint.per-file-ignores]
"__init__.py" = ["E402", "F401"]
"test/*.py" = ["ASYNC"]         # Disable ASYNC check for tests
"*.ipynb" = ["I"]  # Disable import sorting for notebooks

[tool.ruff.lint.mccabe]
max-complexity = 30

[tool.ruff.lint.isort]
combine-as-imports = false

# TODO: move to schema-compatible location
# [toolruff.lint.autoflake]
# in-place = true
# ignore-init-module-imports = true
# remove-all-unused-imports = true
# remove-unused-variables = true
# expand-star-imports = true
# recursive = true

[tool.mypy]
pretty = true
no_implicit_optional = true
namespace_packages = true
show_error_codes = true
python_version = "3.10"
plugins = ["pydantic.mypy"]
exclude = "(^|/)test/data/.*"

[[tool.mypy.overrides]]
module = [
    "jsondiff.*",
    "jsonref.*",
    "jsonschema.*",
    "requests.*",
    "semchunk.*",
    "tabulate.*",
    "transformers.*",
    "tree_sitter_java_orchard.*",
    "yaml.*",
]
ignore_missing_imports = true

[[tool.mypy.overrides]]
module = ["test.*"]
disallow_untyped_defs = false

[tool.semantic_release]
# for default values check:
# https://github.com/python-semantic-release/python-semantic-release/blob/v7.32.2/semantic_release/defaults.cfg

version_source = "tag_only"
branch = "main"

# configure types which should trigger minor and patch version bumps respectively
# (note that they must be a subset of the configured allowed types):
parser_angular_allowed_types = "build,chore,ci,docs,feat,fix,perf,style,refactor,test"
parser_angular_minor_types = "feat"
parser_angular_patch_types = "fix,perf"