pyproject.toml

[project]
name = "data-concierge"
version = "0.1.0"
description = "Verikan — the verified Data Concierge: citation-backed answers about public statistical data, with a reproducible notebook for every answer"
readme = "README.md"
requires-python = ">=3.12"
license = {text = "AGPL-3.0-or-later"}
authors = [
    {name = "datHere"}
]
keywords = ["ai", "open-data", "statistics", "langgraph", "fastapi", "jupyter", "reproducibility"]
classifiers = [
    "Development Status :: 3 - Alpha",
    "Intended Audience :: Science/Research",
    "License :: OSI Approved :: GNU Affero General Public License v3 or later (AGPLv3+)",
    "Programming Language :: Python :: 3.12",
    "Programming Language :: Python :: 3.14",
]

dependencies = [
    # Web Framework
    "fastapi>=0.109.0",
    "uvicorn[standard]>=0.27.0",
    "python-multipart>=0.0.6",
    "pydantic>=2.5.0",
    "pydantic-settings>=2.1.0",

    # Agent Orchestration
    "langgraph>=0.2.0",
    "anthropic>=0.40.0",

    # Data Processing
    "pandas>=2.1.0",
    "numpy>=1.26.0",

    # Vector Store (Qdrant — used for semantic search when configured)

    # Cache
    "redis>=5.0.0",

    # HTTP Client
    "httpx>=0.26.0",

    # qsv runner (timeouts, typed errors) for the onboarding profiling passes
    "qsv-client>=0.4.0",

    # Vector Search (Pinecone for CKAN semantic search)
    # Note: Use the new `pinecone` package (v5+), NOT the legacy
    # `pinecone-client` — the code uses the new Pinecone class and
    # integrated-inference search API which only exist in v3+.
    "pinecone>=5.0.0",

    # Notebook Generation
    "nbformat>=5.9.0",
    "nbclient>=0.10.0",
    "ipykernel>=6.29.0",
    "jinja2>=3.1.0",

    # Configuration
    "python-dotenv>=1.0.0",
    "pyyaml>=6.0.0",

    # Utilities
    "structlog>=24.1.0",
    "tenacity>=8.2.0",

    # Cloud Storage (for persistent storage on Cloud Run)
    "google-cloud-storage>=2.14.0",
]

[project.urls]
Homepage = "https://github.com/dathere/Verikan"
Repository = "https://github.com/dathere/Verikan"
Issues = "https://github.com/dathere/Verikan/issues"

[project.optional-dependencies]
dev = [
    "pytest>=7.4.0",
    "pytest-asyncio>=0.23.0",
    "pytest-cov>=4.1.0",
    "mypy>=1.8.0",
    "ruff>=0.1.0",
    "pre-commit>=3.6.0",
    "httpx>=0.26.0",
    "respx>=0.23.1",
]

[build-system]
requires = ["setuptools>=68.0", "wheel"]
build-backend = "setuptools.build_meta"

[tool.setuptools.packages.find]
where = ["src"]

[tool.ruff]
target-version = "py312"
line-length = 100

[tool.ruff.lint]
select = [
    "E",   # pycodestyle errors
    "W",   # pycodestyle warnings
    "F",   # Pyflakes
    "I",   # isort
    "B",   # flake8-bugbear
    "C4",  # flake8-comprehensions
    "UP",  # pyupgrade
]
ignore = [
    "E501",  # line too long (handled by formatter)
    "B008",  # do not perform function calls in argument defaults
    # UP042 wants `class X(str, Enum)` rewritten to `StrEnum`. That is not a
    # behaviour-preserving change here: with `(str, Enum)`, `str(X.MEMBER)` and
    # f-strings render "X.MEMBER", while StrEnum renders the value. These enums
    # are serialised through Pydantic models and interpolated into prompts,
    # notebooks and logs, so the rewrite would silently change output. Revisit
    # deliberately, with those call sites audited — not as a lint autofix.
    "UP042", # str-enum inheritance; see above
]

[tool.ruff.lint.isort]
known-first-party = ["data_concierge"]

[tool.mypy]
python_version = "3.12"
warn_return_any = true
warn_unused_configs = true
disallow_untyped_defs = true
plugins = ["pydantic.mypy"]

[tool.pytest.ini_options]
asyncio_mode = "auto"
testpaths = ["tests"]
addopts = "-v --cov=src/data_concierge --cov-report=term-missing"

[dependency-groups]
dev = [
    "respx>=0.23.1",
]