pyproject.toml

[build-system]
requires = ["setuptools>=68"]
build-backend = "setuptools.build_meta"

[project]
name = "openextract"
version = "0.1.0"
description = "Self-hosted, API-compatible drop-in replacement for AWS Textract. Runs a local/quantized VLM. ~16-40x cheaper, data stays on your machine."
readme = "README.md"
requires-python = ">=3.10"
license = { text = "Apache-2.0" }
authors = [{ name = "sarcascoder" }]
keywords = ["ocr", "textract", "document-ai", "self-hosted", "local-llm", "vlm"]
classifiers = [
    "License :: OSI Approved :: Apache Software License",
    "Programming Language :: Python :: 3",
    "Topic :: Scientific/Engineering :: Image Recognition",
    "Topic :: Text Processing",
]
dependencies = [
    "fastapi>=0.110",
    "uvicorn>=0.27",
]

[project.optional-dependencies]
tesseract = ["pytesseract>=0.3.10", "pillow>=10"]
pdf = ["pymupdf>=1.24"]
dev = ["pytest>=8", "httpx>=0.27", "boto3>=1.34", "pymupdf>=1.24"]

[project.scripts]
openextract = "openextract.cli:main"

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