pyproject.toml

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

[project]
name = "docling_graph"
version = "1.9.1"
description = "A tool to convert documents into knowledge graphs using Docling."

license = { text = "MIT License" }
keywords = ["docling", "knowledge-graph", "nlp", "pdf", "graph"]
classifiers = [
    "Operating System :: MacOS :: MacOS X",
    "Operating System :: POSIX :: Linux",
    "Operating System :: Microsoft :: Windows",
    "Intended Audience :: Developers",
    "Programming Language :: Python :: 3",
    "Programming Language :: Python :: 3.10",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Programming Language :: Python :: 3.13",    
]
readme = "README.md"
authors = [
  { name = "Ayoub El Bouchtili", email = "ayoub.elbouchtili@fr.ibm.com" },
  { name = "Michele Dolfi", email = "dol@zurich.ibm.com" },
  { name = "Maxime Gillot", email = "Maxime.Gillot@ibm.com" },
  { name = "Sophie Lang", email = "sophie.lang@de.ibm.com" },
  { name = "Guilhaume Leroy Meline", email = "guilhaume@fr.ibm.com" },
  { name = "Peter Staar", email = "taa@zurich.ibm.com" },
]
requires-python = ">=3.10,<4.0"
dependencies = [
    # DocLang input/export + serializer surface requires docling>=2.100 /
    # docling-core>=2.86. The docling.service_client SDK used for remote
    # docling-serve conversion needs >=2.104 (aligned convert(source=) API;
    # all imported exceptions exist since 2.102); floor set at 2.105.0,
    # whose service-client surface is identical to the locked version and
    # carries the multipart nested-options encoding fix (#3672).
    "docling>=2.105.0,<3.0.0",
    "docling-core[chunking, chunking-openai]>=2.86.0,<3.0.0",
    "pydantic>=2.0.0,<3.0.0",
    "networkx>=3.0.0,<4.0.0",
    "rich>=13,<16",
    "typer[all]>=0.12,<1.0.0",
    "python-dotenv>=1.0,<2.0",
    "litellm>=1.0.0,<2.0.0",
    "pyyaml>=6.0,<7.0",
    "aiofiles>=24.0.0,<26.0.0",
    "tenacity>=9.1.4",
    # URL input handling (docling-serve conversion uses docling.service_client)
    "requests>=2.31,<3.0",
]

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

[project.scripts]
docling-graph = "docling_graph.cli.main:main"

[project.optional-dependencies]
vlm = [
    "docling[vlm]>=2.105.0,<3.0.0",
]
# Legacy provider SDKs removed; LiteLLM is installed by default.
# Amazon Bedrock routes through LiteLLM but needs boto3 for the AWS
# credential chain and SigV4 signing.
bedrock = [
    "boto3>=1.34.0,<2.0.0",
]
semantic = [
    "rapidfuzz",
]
# Ontology-based template generation (docling-graph template from-ontology):
# rdflib for OWL/RDFS/SKOS, linkml-runtime for LinkML SchemaView. The JSON
# Schema compiler is stdlib-only and needs neither.
templategen = [
    "rdflib>=7.0,<8.0",
    "linkml-runtime>=1.7,<2.0",
]







[dependency-groups]
# Development tools (linters, type-checkers) and utilities
dev = [
    "pre-commit~=4.5",
    "mypy~=1.10",
    "types-setuptools>=80.9,<84.0",
    "pandas-stubs~=2.1",
    "types-openpyxl~=3.1",
    "types-requests~=2.31",
    "boto3-stubs~=1.37",
    "types-urllib3~=1.26",
    "types-tqdm~=4.67",
    "coverage~=7.6",
    "pytest~=9.0",
    "pytest-cov>=6.1.1",
    "pytest-dependency~=0.6",
    "pytest-durations>=1.6.1,<1.8.0",
    "pytest-xdist~=3.3",
    "ipykernel~=7.1",
    "nbqa~=1.9",
    "python-semantic-release>=9.0.0",
    "mypy>=1.4.0",
    "ruff>=0.0.280",
    # Documentation tools
    "mkdocs>=1.5.0",
    "mkdocs-material>=9.0.0",
    "mkdocstrings[python]>=0.24.0",
    "pymdown-extensions>=10.0.0",
]

# Test stack and utilities
test = [
    "pytest>=7.4.0",
    "pytest-cov>=4.1.0",
    "pytest-mock>=3.11.0",
    "pytest-xdist>=3.3.0",
    "pytest-timeout>=2.1.0",
    "hypothesis>=6.82.0",
    "pytest-asyncio>=0.21.0",
    "freezegun>=1.2.2",
    "responses>=0.23.0",
    "faker>=19.0.0",
    "types-PyYAML",
    "types-requests",
    "mypy>=1.4.0",
    # Cypher exporter integration tests (skip cleanly when Docker is absent)
    "neo4j>=5.20.0",
    "testcontainers>=4.8.0",
]

# Notebook-related tools and data analysis helpers
notebook = [
    "notebook",
    "ipython",
    "pandas",
]

[tool.uv]
package = true
default-groups = ["dev", "test", "notebook"]

[tool.setuptools.packages.find]
include = ["docling_graph*"]

[tool.ruff]
target-version = "py310"
line-length = 100
respect-gitignore = true

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

[tool.ruff.lint]
select = [
  "E",    # pycodestyle errors
  "W",    # pycodestyle warnings
  "F",    # pyflakes
  "I",    # isort
  "N",    # pep8-naming
  "UP",   # pyupgrade
  "B",    # flake8-bugbear
  "A",    # flake8-builtins
  "C4",   # flake8-comprehensions
  "C90",  # mccabe (complexity)
  "ANN",  # flake8-annotations
  "PD",   # pandas-vet
  "PIE",  # flake8-pie
  "Q",    # flake8-quotes
  "RUF",  # Ruff-specific checks
  "ASYNC" # async
]
ignore = [
  "C408",   # Unnecessary `dict()` call (rewrite as a literal)
  "E501",   # Line too long, handled by ruff formatter
  "D107",   # "Missing docstring in __init__",
  "F401",   # imported but unused; consider using `importlib.util.find_spec` to test for "
  "F811",   # "redefinition of the same function"
  "PL",     # Pylint
  "RUF012", # Mutable Class Attributes
  "UP006",  # List vs list, etc
  "UP007",  # Option and Union
  "UP035",  # `typing.Set` is deprecated, use `set` instead"
  # "ANN101",  # Missing type annotation for self
  # "ANN102",  # Missing type annotation for cls
  "ANN401",  # Dynamically typed expressions (Any)
]

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

[tool.ruff.lint.per-file-ignores]
"__init__.py" = ["E402", "F401"]
"tests/**/*.py" = ["ASYNC", "ANN201", "ANN001"]
# Rendered templategen goldens keep the schema-definition docs' Optional[...]
# idiom (template-basics.md required imports) instead of `X | None`.
"tests/fixtures/templategen/golden/*.py" = ["UP045"]

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

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

[tool.mypy]
pretty = true
python_version = "3.12"
no_implicit_optional = true
exclude = ["^site/", "^docs/examples/"]
plugins = "pydantic.mypy"
warn_return_any = true
warn_unused_configs = true
disallow_untyped_defs = true
check_untyped_defs = true
warn_redundant_casts = true
warn_unused_ignores = true
warn_no_return = true
warn_unreachable = true
strict_optional = true
strict_equality = true

[[tool.mypy.overrides]]
module = [
    "docling.*",
    "rich.*",
    "typer.*",
    "yaml.*",
    "pandas.*",
    "networkx.*",
    "litellm.*",
    "tiktoken.*",
    "google.*",
    "rapidfuzz",
]
ignore_missing_imports = true

[tool.semantic_release]
version_toml = ["pyproject.toml:project.version"]
version_variables = ["docling_graph/__init__.py:__version__"]
commit_message = "chore(release): bump package version to {version}"

# Prevent automatic major version bumps - major releases require manual Git tags
major_on_zero = false
allow_zero_version = false

[tool.semantic_release.branches.main]
match = "main"
prerelease = false

# Custom commit parser to cap BREAKING CHANGE at minor releases
[tool.semantic_release.commit_parser_options]
allowed_tags = ["feat", "fix", "perf", "refactor", "docs", "style", "test", "build", "ci", "chore"]
minor_tags = ["feat"]
patch_tags = ["fix", "perf", "refactor"]

[tool.semantic_release.changelog]
# Update mode: insert new release after insertion_flag (see CHANGELOG.md).
mode = "update"
insertion_flag = "<!-- version list -->"
exclude_commit_patterns = [
  "chore\\(release\\):.*",  # exclude version-bump commits from changelog
]

[tool.semantic_release.changelog.default_templates]
changelog_file = "CHANGELOG.md"

[tool.semantic_release.changelog.environment]
block_start_string = "{%"
block_end_string = "%}"
variable_start_string = "{{"
variable_end_string = "}}"
comment_start_string = "{#"
comment_end_string = "#}"
trim_blocks = false
lstrip_blocks = false
newline_sequence = "\n"
keep_trailing_newline = false
extensions = []
autoescape = true

[tool.semantic_release.remote]
name = "origin"
type = "github"
ignore_token_for_push = false

[tool.semantic_release.publish]
dist_glob_patterns = ["dist/*"]
upload_to_vcs_release = true

[tool.pytest.ini_options]
filterwarnings = [
    "error",
    # Ignore ResourceWarnings from unclosed sockets/event loops
    "ignore::ResourceWarning",
    # Ignore warnings from external dependencies
    "ignore:.*Use BaseMetaSerializer.*:DeprecationWarning",
    "ignore:.*Using `@model_validator` with mode='after' on a classmethod is deprecated.*:pydantic.warnings.PydanticDeprecatedSince212",
]
addopts = "--cov=docling_graph --cov-report=term-missing"
testpaths = ["tests"]

[tool.coverage.run]
source = ["docling_graph"]
omit = ["tests/*"]

[tool.coverage.report]
exclude_lines = [
    "pragma: no cover",
    "def __repr__",
    "if __name__ == .__main__.:",
    "raise NotImplementedError",
]