pyproject.toml

# ContextSafe - Document Anonymization System
# ============================================

[build-system]
requires = ["poetry-core>=1.0.0"]
build-backend = "poetry.core.masonry.api"

[tool.poetry]
name = "contextsafe"
version = "0.1.0"
description = "Sistema 100% local de anonimizacion inteligente de documentos sensibles"
authors = ["ContextSafe Team"]
readme = "README.md"
license = "Apache-2.0"
packages = [{ include = "contextsafe", from = "src" }]

[tool.poetry.dependencies]
python = "^3.11"

# Core - Design by Contract
icontract = "^2.6.0"

# Data Validation
pydantic = "^2.0.0"
pydantic-settings = "^2.0.0"

# Structured Logging
structlog = "^23.0.0"

# HTTP Client
httpx = "^0.24.0"

# Web Framework
fastapi = "^0.100.0"
uvicorn = { version = "^0.23.0", extras = ["standard"] }

# Database - SQLite with encryption
sqlalchemy = { version = "^2.0.0", extras = ["asyncio"] }
aiosqlite = "^0.19.0"
sqlcipher3-binary = "^0.5.0"

# Document Processing
pdfplumber = "^0.10.0"
python-docx = "^0.8.11"
chardet = "^5.0.0"
reportlab = "^4.0.0"

# OCR
pytesseract = "^0.3.10"
Pillow = "^10.0.0"
opencv-python-headless = "^4.8.0"

# NLP / NER
spacy = "^3.7.0"
transformers = "^4.36.0"
torch = "^2.1.0"

# Fuzzy String Matching (for entity normalization)
thefuzz = "^0.22.1"

# LLM Runtime (legacy - puede eliminarse si no se usa)
# llama-cpp-python = "^0.2.50"

# Embeddings (optional, for semantic alias)
sentence-transformers = "^2.2.0"

# Metrics
prometheus-client = "^0.17.0"

# WebSocket
websockets = "^12.0"

# CLI
click = "^8.1.0"

# PII Detection (Presidio)
presidio-analyzer = "^2.2.0"
presidio-anonymizer = "^2.2.0"

python-dotenv = "^1.0.0"
python-multipart = "^0.0.28"

[tool.poetry.group.dev.dependencies]
# Testing
pytest = "^7.4.0"
pytest-asyncio = "^0.21.0"
pytest-cov = "^4.1.0"
hypothesis = "^6.82.0"

# Type Checking
mypy = "^1.5.0"

# Linting & Formatting
ruff = "^0.1.0"

# Security Scanning
bandit = "^1.7.0"

# Mutation Testing
mutmut = "^2.4.0"

# E2E Testing (Web)
pytest-playwright = "^0.4.0"

# Test Utilities
factory-boy = "^3.3.0"
faker = "^19.0.0"
testcontainers = "^3.7.0"

[tool.poetry.group.docs.dependencies]
mkdocs = "^1.5.0"
mkdocs-material = "^9.0.0"
mkdocstrings = { version = "^0.23.0", extras = ["python"] }

[tool.poetry.scripts]
contextsafe = "contextsafe.api.main:run"

# ============================================
# TOOL CONFIGURATIONS
# ============================================

[tool.ruff]
target-version = "py311"
line-length = 100
src = ["src", "tests"]

[tool.ruff.lint]
select = [
    "E",      # pycodestyle errors
    "W",      # pycodestyle warnings
    "F",      # Pyflakes
    "I",      # isort
    "C",      # flake8-comprehensions
    "B",      # flake8-bugbear
    "UP",     # pyupgrade
    "S",      # flake8-bandit
    "A",      # flake8-builtins
    "C4",     # flake8-comprehensions
    "DTZ",    # flake8-datetimez
    "T10",    # flake8-debugger
    "EXE",    # flake8-executable
    "ISC",    # flake8-implicit-str-concat
    "ICN",    # flake8-import-conventions
    "G",      # flake8-logging-format
    "INP",    # flake8-no-pep420
    "PIE",    # flake8-pie
    "PYI",    # flake8-pyi
    "PT",     # flake8-pytest-style
    "Q",      # flake8-quotes
    "RSE",    # flake8-raise
    "RET",    # flake8-return
    "SLF",    # flake8-self
    "SLOT",   # flake8-slots
    "SIM",    # flake8-simplify
    "TID",    # flake8-tidy-imports
    "TCH",    # flake8-type-checking
    "ARG",    # flake8-unused-arguments
    "PTH",    # flake8-use-pathlib
    "PL",     # Pylint
    "TRY",    # tryceratops
    "FLY",    # flynt
    "PERF",   # Perflint
    "RUF",    # Ruff-specific rules
]
ignore = [
    "S101",    # Use of assert (needed for icontract)
    "S104",    # Possible binding to all interfaces
    "B008",    # Do not perform function calls in argument defaults (FastAPI Depends)
    "PLR0913", # Too many arguments to function call
    "PLR2004", # Magic value used in comparison
    "UP007",   # Use X | Y instead of Optional[X] (not yet adopted)
    "G004",    # Logging statement uses f-string
    "TCH001",  # Move application import into type-checking block
    "TCH002",  # Move third-party import into type-checking block
    "TCH003",  # Move standard library import into type-checking block
    "TCH005",  # Empty type-checking block
    "S311",    # Standard pseudo-random generators not suitable for cryptography
    "RUF012",  # Mutable class attributes should be annotated with ClassVar
    "RUF001",  # String contains ambiguous character (Spanish text)
    "RUF003",  # Comment contains ambiguous character (Spanish comments)
    "TRY003",  # Avoid specifying long messages outside the exception class
    "TRY300",  # Consider moving statement to else block
    "ARG002",  # Unused method argument
    "C901",    # Function is too complex
    "SLF001",  # Private member accessed
    "DTZ003",  # datetime.utcnow() used
    "PLR0912", # Too many branches
    "SIM102",  # Use a single if statement instead of nested if
    "RET505",  # Unnecessary else after return
    "PTH123",  # open() should be replaced with Path.open()
    "E501",    # Line too long (handled by formatter)
    "ISC001",  # Implicit string concatenation (conflicts with formatter)
    "TRY400",  # Use logging.exception instead of logging.error
    "TRY200",  # Use raise from to specify exception cause
    "PLW0603", # Using global statement
    "PLR0911", # Too many return statements
    "PLR0915", # Too many statements
    "PLW2901", # Loop variable overwritten by assignment
    "S110",    # try/except/pass detected
    "S603",    # subprocess call: check for execution of untrusted input
    "S607",    # Starting a process with a partial executable path
    "ARG001",  # Unused function argument
    "RET504",  # Unnecessary assignment before return
    "PERF401", # Use list comprehension (not always clearer)
    "DTZ001",  # datetime without timezone info
    "RUF013",  # Implicit Optional
    "RUF002",  # Docstring contains ambiguous character (Spanish)
    "PT011",   # pytest.raises too broad
    "B904",    # Use raise from inside except
    "B017",    # pytest.raises(Exception) considered evil
    "B033",    # Set contains duplicate items
    "B007",    # Loop variable not used in loop body
    "PT004",   # Fixture does not return anything, use underscore prefix
    "A002",    # Argument shadowing builtin
    "A001",    # Variable shadowing builtin
    "SIM110",  # Use any()/all() instead of for loop
    "PYI034",  # __new__ should return self
    "PIE810",  # Pass multiple values to isinstance
    "TRY301",  # Abstract raise to inner function
    "TRY201",  # Use raise without args in except
    "TID252",  # Prefer relative imports
    "PTH109",  # os.getcwd() -> Path.cwd()
    "PTH108",  # os.unlink() -> Path.unlink()
    "PLR5501", # Use elif instead of else+if
    "C401",    # Unnecessary generator passed to set()
]

[tool.ruff.lint.per-file-ignores]
"tests/**/*.py" = [
    "S101",   # asserts allowed in tests
    "ARG",    # Unused function args (fixtures)
    "S105",   # Possible hardcoded password
    "S106",   # Possible hardcoded password
]

[tool.ruff.lint.isort]
known-first-party = ["contextsafe"]
force-single-line = false
lines-after-imports = 2

[tool.mypy]
python_version = "3.11"
warn_return_any = true
warn_unused_ignores = true
disallow_untyped_defs = true
disallow_incomplete_defs = true
check_untyped_defs = true
no_implicit_optional = true
warn_redundant_casts = true
plugins = ["pydantic.mypy"]
explicit_package_bases = true
mypy_path = "src"
disable_error_code = [
    "import-untyped",
    "misc",
    "type-arg",
    "list-item",
    "no-any-return",
    "no-untyped-def",
    "no-untyped-call",
    "arg-type",
    "union-attr",
    "assignment",
    "override",
    "call-arg",
    "valid-type",
    "return-value",
    "index",
    "call-overload",
    "attr-defined",
    "operator",
    "unused-ignore",
    "import-not-found",
]

[[tool.mypy.overrides]]
module = [
    "pdfplumber.*",
    "docx.*",
    "pytesseract.*",
    "cv2.*",
    "spacy.*",
    "llama_cpp.*",
    "sentence_transformers.*",
    "sqlcipher3.*",
    "prometheus_client.*",
    "icontract.*",
    "icontract_hypothesis.*",
    "hypothesis.*",
    "factory.*",
    "faker.*",
    "testcontainers.*",
]
ignore_missing_imports = true

[tool.pytest.ini_options]
asyncio_mode = "auto"
testpaths = ["tests"]
python_files = ["test_*.py"]
python_classes = ["Test*"]
python_functions = ["test_*"]
addopts = [
    "-v",
    "--strict-markers",
    "--tb=short",
    "-ra",
]
markers = [
    "unit: Unit tests",
    "integration: Integration tests",
    "contract: Contract tests",
    "pbt: Property-based tests",
    "e2e: End-to-end tests",
    "slow: Slow tests",
]
filterwarnings = [
    "ignore::DeprecationWarning",
]

[tool.coverage.run]
source = ["src/contextsafe"]
branch = true
omit = [
    "*/tests/*",
    "*/__init__.py",
]

[tool.coverage.report]
exclude_lines = [
    "pragma: no cover",
    "def __repr__",
    "raise NotImplementedError",
    "if TYPE_CHECKING:",
    "if __name__ == .__main__.:",
    "@abstractmethod",
    "Protocol",
]
fail_under = 0
show_missing = true

[tool.bandit]
exclude_dirs = ["tests", "scripts"]
skips = [
    "B101",  # assert_used - needed for icontract
    "B104",  # hardcoded_bind_all_interfaces - accepted, configured via env
    "B110",  # try_except_pass - accepted pattern for optional features
    "B311",  # random - not used for cryptography, only for synthetic data
    "B404",  # subprocess import - legitimate use for NLP tools
    "B603",  # subprocess_without_shell_equals_true - commands are controlled
    "B607",  # start_process_with_partial_path - commands are controlled
]