pyproject.toml

[project]
name = "gsma-dataset-creation"
version = "0.1.1"
description = "GSMA document processing pipeline"
readme = "README.md"
requires-python = ">=3.11"
dependencies = [
    "python-docx>=1.1.0",
    "pymupdf[all]>=1.24.0",
    "loguru>=0.7.2",
    "typer>=0.12.0",
    "tqdm>=4.66.0",
    "chonkie[semantic,st]>=0.5.1",
    "requests>=2.28.0",
    "numpy>=2.3.3",
    "openai>=1.107.2",
    "pydantic>=2.11.7",
    "python-dotenv>=1.1.1",
    "torch>=2.8.0",
    "sentence-transformers>=5.1.0,<5.1.1",
    "datasets>=4.1.0",
    "accelerate>=0.26.0",
    "aiosqlite>=0.20.0",
    "dvc-s3[dev]>=3.2.2",
    "tenacity>=8.2.0",
    "tiktoken>=0.11.0",
    "json-repair>=0.51.0",
    "unidecode>=1.3.8",
    "langdetect>=1.0.9",
    "faiss-cpu>=1.12.0",
    "pandas>=2.3.2",
    "pyarrow>=21.0.0",
    "scipy>=1.16.2",
    "playwright>=1.40.0",
    "httpx>=0.25.0",
    "lxml>=6.0.2",
    "levenshtein>=0.27.1",
    "tabulate>=0.9.0",
]

[project.scripts]
gsma = "gsma_dataset_creation.cli:app"

[project.optional-dependencies]
dev = [
    "dvc>=3.63.0",
    "pre-commit>=3.0.0",
    "ruff>=0.1.0",
    "bandit>=1.8.6",
    "ipython>=9.6.0",
    "argilla>=2.0.0",
]
test = [
    "pytest>=8.0.0",
    "pytest-asyncio>=0.21.0",
    "pytest-cov>=4.0.0",
    "coverage>=7.0.0",
    "hypothesis>=6.90.0",
    "mypy>=1.8.0",
    "types-tqdm>=4.67.0",
]
filters = [
    "setfit>=1.1.3",
]

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

[tool.setuptools.packages.find]
include = ["gsma_dataset_creation*"]
exclude = ["tests*", "data*"]

[tool.bandit]
# Only check for hardcoded credentials/passwords in specific directories
tests = ["B105", "B106", "B107"]
targets = ["gsma_dataset_creation", "scripts", "tests"]

[tool.coverage.run]
omit = [
    "*/tests/*",
    "**/__main__.py",
    "**/__init__.py",
]

[tool.coverage.report]
exclude_lines = [
    "pragma: no cover",
    "def __repr__",
    "if self.debug:",
    "if settings.DEBUG",
    "raise AssertionError",
    "raise NotImplementedError",
    "if 0:",
    "if __name__ == .__main__.:",
    "class .*\\bProtocol\\):",
    "@(abc\\.)?abstractmethod",
]

[tool.pytest.ini_options]
asyncio_mode = "auto"

[tool.mypy]
strict = true
warn_return_any = true
warn_unused_configs = true
disallow_untyped_defs = true