pyproject.toml

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

[project]
dependencies = [
    "numpy",
    "pandas",
    "seaborn",
    "jupyter",  # to run jupyter notebooks.
    "python-magic-bin",  # to detect file type of files.
    "oletools[full]",  # to analyze rtf and office files.
    "sympy",
    "pydantic<2.0",  # to manage dataclasses.
    "requests",  # to send api calls.
    "python-dotenv",  # to load .env files
    "ratelimit",  # to limit api calls.
    "textract>=1.6.4",  # to extract text from files.
    "pywin32",  # for windows metadata extraction.
    "python-docx",  # to read word files
    "python-docx-docm",  # to read word files with marcros files
    "langdetect",  # to detect language
    "pypdf[full]",  # read pdf documents
    "filetype",  # for second opinion on file type.
    "rarfile",  # for extraction of rar files.
    "transformers~=4.27.1",  # to load huggingface models.
    "langchain",  # to use llms in text analysis.
    "llama_index==0.7.4",  # to index documents for llms.
    "sentence_transformers",
    "protobuf",
    "cpm_kernels",  # for ChatGLM models
    "pyarrow~=12.0.1",
    "textsplit~=0.5",  # for splitting text into embeddings.
    "ray[default]~=2.6.3",
    "ray[tune]~=2.6.3",
    "striprtf~=0.0.26"
]
description = "Clustering of malware documents."
license = {text = "unspecified"}
name = "malwaredoc"
readme = "README.md"
requires-python = ">=3.10"
version = "0.0.1"

[project.optional-dependencies]
dev = [
    "pre-commit"
]

[tool.black]
experimental-string-processing = true
include = '\.pyi?$'
line-length = 120
target-version = ['py39', 'py310', 'py311']

[tool.isort]
profile = "black"

[tool.pytest.ini_options]
addopts = [
    "--import-mode=importlib"
]

[tool.ruff]
# Allow unused variables when underscore-prefixed.
dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"
# Exclude a variety of commonly ignored directories.
exclude = [
    ".bzr",
    ".direnv",
    ".eggs",
    ".git",
    ".git-rewrite",
    ".hg",
    ".mypy_cache",
    ".nox",
    ".pants.d",
    ".pytype",
    ".ruff_cache",
    ".svn",
    ".tox",
    ".venv",
    "__pypackages__",
    "_build",
    "buck-out",
    "build",
    "dist",
    "node_modules",
    "venv"
]
# Allow autofix for all enabled rules (when `--fix`) is provided.
fixable = ["ALL"]
ignore = ["E402"]
# Same as Black.
line-length = 120
per-file-ignores = {}
# Enable the pycodestyle (`E`) and Pyflakes (`F`) rules by default.
# Unlike Flake8, Ruff doesn't enable pycodestyle warnings (`W`) or
# McCabe complexity (`C901`) by default.
select = ["E", "F"]
# Assume Python 3.10.
target-version = "py310"
unfixable = []

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