pyproject.toml

# OpenProcessor - Python Project Configuration
# Configuration for ruff (linter + formatter) and project metadata

[project]
name = "openprocessor"
version = "0.1.0"
description = "Self-hosted AI-powered visual processing engine with object detection, face recognition, CLIP embeddings, OCR, and vector search"
readme = "README.md"
requires-python = ">=3.12"
authors = [
    {name = "OpenProcessor Contributors"}
]
license = {text = "MIT"}

classifiers = [
    "Development Status :: 5 - Production/Stable",
    "Intended Audience :: Developers",
    "Topic :: Scientific/Engineering :: Artificial Intelligence",
    "Topic :: Scientific/Engineering :: Image Recognition",
    "License :: OSI Approved :: MIT License",
    "Programming Language :: Python :: 3.12",
]

keywords = ["yolo", "triton", "tensorrt", "object-detection", "face-recognition", "clip", "ocr", "visual-search", "fastapi", "inference", "openprocessor"]

dependencies = [
    "fastapi",
    "uvicorn[standard]",
    "python-multipart",
    "orjson",
    "pydantic-settings",
    "ultralytics",
    "tritonclient[all]",
    "onnx>=1.12.0,<=1.19.1",
    "onnxslim>=0.1.71",
    "onnxsim>=0.4.33",
    "onnxscript",
    "onnxruntime-gpu",
    "onnx-graphsurgeon",
    "tensorrt-cu12==10.13.3.9",
    "nvidia-dali-cuda120",
    "torch",
    "torchvision",
    "opencv-python",
    "numpy",
    "scipy",
    "requests",
    "aiohttp",
    "matplotlib",
    "seaborn",
    "pandas",
    "psutil",
    "tqdm",
    "tabulate",
    "pyyaml",
    "opensearch-py>=2.3.0",
    "transformers>=4.30.0",
    "timm",
    "huggingface_hub",
]

[project.optional-dependencies]
dev = [
    "pre-commit>=4.0.0",
    "ruff>=0.8.4",
    "mypy>=1.13.0",
    "bandit[toml]>=1.8.0",
    "pytest>=8.0.0",
    "pytest-cov>=5.0.0",
    "pytest-asyncio>=0.24.0",
]

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

# =============================================================================
# Ruff Configuration (Linter + Formatter)
# =============================================================================

[tool.ruff]
# Target Python 3.12
target-version = "py312"

# Line length (industry standard: 88-100, we use 100)
line-length = 100

# Directories to include/exclude
include = ["src/**/*.py", "export/**/*.py", "dali/**/*.py", "scripts/**/*.py", "tests/**/*.py"]
exclude = [
    ".git",
    ".venv",
    "__pycache__",
    ".pytest_cache",
    ".mypy_cache",
    ".ruff_cache",
    "reference_repos",
    "pytorch_models",
    "models",
    "cache",
    "benchmarks/results",
    "*.egg-info",
]

# =============================================================================
# Ruff Linter Rules
# =============================================================================

[tool.ruff.lint]
# Enable rule categories (industry standard set)
select = [
    "E",     # pycodestyle errors
    "W",     # pycodestyle warnings
    "F",     # pyflakes
    "I",     # isort (import sorting)
    "N",     # pep8-naming
    "UP",    # pyupgrade (modernize Python code)
    "B",     # flake8-bugbear (detect common bugs)
    "A",     # flake8-builtins (shadowing builtins)
    "C4",    # flake8-comprehensions (better list/dict comprehensions)
    "DTZ",   # flake8-datetimez (timezone-aware datetime)
    "T10",   # flake8-debugger (no breakpoints in production)
    "EM",    # flake8-errmsg (better error messages)
    "ISC",   # flake8-implicit-str-concat
    "ICN",   # flake8-import-conventions
    "PIE",   # flake8-pie (misc lints)
    "PT",    # flake8-pytest-style
    "Q",     # flake8-quotes (enforce quote style)
    "RSE",   # flake8-raise (better exception raising)
    "RET",   # flake8-return (better return statements)
    "SIM",   # flake8-simplify (simplify code)
    "TID",   # flake8-tidy-imports
    "TCH",   # flake8-type-checking (optimize TYPE_CHECKING imports)
    "ARG",   # flake8-unused-arguments
    "PTH",   # flake8-use-pathlib (prefer pathlib over os.path)
    "ERA",   # eradicate (remove commented-out code)
    "PL",    # pylint
    "PERF",  # perflint (performance anti-patterns)
    "RUF",   # ruff-specific rules
]

# Rules to ignore (with justification)
ignore = [
    "E501",    # Line too long (handled by formatter)
    "PLC0415", # Import should be at top-level (lazy imports for heavy deps)
    "PLR0913", # Too many arguments (common in ML code)
    "PLR0912", # Too many branches (common in validation logic)
    "PLR0915", # Too many statements (common in setup functions)
    "PLR2004", # Magic value comparison (common with thresholds)
    "N803",    # Argument name should be lowercase (allow numpy-style names)
    "N806",    # Variable should be lowercase (allow numpy-style names)
    "DTZ005",  # datetime.now() without tz (not critical for logs)
    "EM101",   # Exception must not use string literal (too strict)
    "EM102",   # Exception must not use f-string (too strict)
    "TRY003",  # Avoid long messages in exception (too strict)
    "SIM108",  # Use ternary operator (readability preference)
    "PTH123",  # pathlib open() vs builtin open() (performance reasons)
    "RUF012",  # Mutable class attributes (common in Pydantic)
]

# Allow autofix for all enabled rules
fixable = ["ALL"]
unfixable = []

# Allow unused variables when prefixed with underscore
dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"

[tool.ruff.lint.per-file-ignores]
# Test files can have additional flexibility
"tests/**/*.py" = [
    "ARG001",  # Unused function argument (fixtures)
    "S101",    # Use of assert
    "PLR2004", # Magic values
    "PT028",   # Default arguments in test-like functions (standalone test scripts, not pytest)
    "RUF001",  # Ambiguous Unicode characters (intentional for display)
]

# Scripts can be less strict
"scripts/**/*.py" = [
    "T201",    # print() allowed in scripts
    "PLR2004", # Magic values
    "PT028",   # Default arguments in test-named functions (these are scripts, not pytest tests)
]

# DALI pipeline scripts are not pytest tests despite test_* naming
"dali/**/*.py" = [
    "PT028",   # Default arguments in test-like functions (standalone validation scripts)
    "ARG001",  # Unused function argument (kept for interface consistency)
    "ERA001",  # Commented-out code (documentation comments describing matrix format)
    "RUF022",  # Unsorted __all__ (intentionally grouped by category)
]

# FastAPI routers use dependency injection in default arguments (standard pattern)
"src/routers/**/*.py" = [
    "B008",    # Function call in default argument (FastAPI File(), Depends(), Query())
    "PT028",   # Default arguments in endpoint functions (FastAPI pattern, not pytest)
]

# Utils package has documentation comments explaining import structure
"src/utils/__init__.py" = [
    "ERA001",  # Commented-out code (import documentation for circular dependency prevention)
]

# Ultralytics patches are third-party code
"src/ultralytics_patches/**/*.py" = [
    "ALL",     # Disable all linting for third-party patches
]

# Export scripts for TensorRT plugins (third-party TRT NMS patterns)
"export/**/*.py" = [
    "N801",    # Class names follow TensorRT plugin conventions (TRT_EfficientNMS)
    "ARG001",  # Unused arguments kept for interface consistency
    "ARG004",  # TensorRT plugin forward() requires all args for ONNX export
    "PT028",   # Default arguments in test-named functions (standalone scripts, not pytest)
    "F401",    # Import for side effects (pycuda.autoinit, pycuda.driver)
]

[tool.ruff.lint.isort]
# Import sorting configuration (follows black/blue style)
force-single-line = false
lines-after-imports = 2
known-first-party = ["src"]
known-third-party = ["ultralytics", "tritonclient", "fastapi", "pydantic"]
section-order = ["future", "standard-library", "third-party", "first-party", "local-folder"]

# Combine 'from' imports from the same module
combine-as-imports = true
split-on-trailing-comma = false

[tool.ruff.lint.flake8-quotes]
# Use single quotes (blue style, more compact)
inline-quotes = "single"
multiline-quotes = "double"
docstring-quotes = "double"

[tool.ruff.lint.pydocstyle]
# Use Google-style docstrings
convention = "google"

[tool.ruff.lint.pylint]
# Pylint-specific settings
max-args = 10
max-branches = 15
max-returns = 8
max-statements = 60

# =============================================================================
# Ruff Formatter Configuration
# =============================================================================

[tool.ruff.format]
# Use single quotes (blue style)
quote-style = "single"

# Indent with spaces
indent-style = "space"

# Respect existing line endings
line-ending = "auto"

# Enable magic trailing comma (helps with git diffs)
skip-magic-trailing-comma = false

# Docstring formatting
docstring-code-format = true
docstring-code-line-length = 88

# =============================================================================
# Pytest Configuration
# =============================================================================

[tool.pytest.ini_options]
testpaths = ["tests"]
python_files = ["test_*.py", "*_test.py"]
python_classes = ["Test*"]
python_functions = ["test_*"]
addopts = [
    "-v",
    "--strict-markers",
    "--tb=short",
    "--cov=src",
    "--cov-report=term-missing",
    "--cov-report=html",
]

# =============================================================================
# Coverage Configuration
# =============================================================================

[tool.coverage.run]
source = ["src"]
omit = [
    "*/tests/*",
    "*/test_*.py",
    "*/__pycache__/*",
    "*/site-packages/*",
    "src/ultralytics_patches/*",  # Third-party code
]

[tool.coverage.report]
precision = 2
show_missing = true
skip_covered = false
exclude_lines = [
    "pragma: no cover",
    "def __repr__",
    "raise AssertionError",
    "raise NotImplementedError",
    "if __name__ == .__main__.:",
    "if TYPE_CHECKING:",
    "@abstractmethod",
]

# =============================================================================
# MyPy Configuration (Type Checking)
# =============================================================================

[tool.mypy]
python_version = "3.12"
warn_return_any = false  # Disabled for ML code with numpy/FAISS operations
warn_unused_configs = true
disallow_untyped_defs = false
disallow_incomplete_defs = false
check_untyped_defs = true
no_implicit_optional = true
warn_redundant_casts = true
warn_unused_ignores = false
warn_no_return = true
strict_equality = true
ignore_missing_imports = true
explicit_package_bases = true
mypy_path = "."
namespace_packages = true

# Ignore third-party packages without type stubs
[[tool.mypy.overrides]]
module = [
    "ultralytics.*",
    "tritonclient.*",
    "onnx.*",
    "onnxslim.*",
    "onnxsim.*",
    "onnxruntime.*",
    "tensorrt.*",
    "nvidia.dali.*",
    "cv2.*",
    "opensearchpy.*",
    "transformers.*",
    "timm.*",
]
ignore_missing_imports = true

# Relax strict checks for ML/scientific code with FAISS and numpy operations
[[tool.mypy.overrides]]
module = [
    "src.services.clustering",
    "src.services.duplicate_detection",
    "src.clients.opensearch",
    "src.clients.triton_client",
    "src.services.model_export",
    "src.services.visual_search",
    "src.utils.retry",
    "src.utils.face_alignment",
    "src.utils.image_processing",
    "src.services.image",
    "src.services.face_identity",
    "src.routers.detect",
]
disable_error_code = ["attr-defined", "operator", "misc", "assignment", "return-value", "call-overload", "union-attr", "arg-type", "dict-item"]

# Skip type checking on third-party patches
[[tool.mypy.overrides]]
module = "src.ultralytics_patches.*"
ignore_errors = true

# =============================================================================
# Bandit Configuration (Security Linting)
# =============================================================================

[tool.bandit]
exclude_dirs = [
    "tests",
    ".venv",
    "venv",
    "reference_repos",
    "pytorch_models",
    "cache",
]
skips = [
    "B101",  # assert_used - common in test files
    "B311",  # random - only used for non-cryptographic retry jitter
    "B601",  # paramiko_calls - not used
]

[tool.bandit.assert_used]
skips = ["*_test.py", "test_*.py"]