pyproject.toml

[build-system]
build-backend = "setuptools.build_meta"
# Pin to 62.6 for support from reading requirements from requirements.txt
requires = ["setuptools >= 62.6.0"]

[project]
authors = [
    {email = "white.d.andrew@gmail.com", name = "Andrew White"},
]
# Full list: https://pypi.python.org/pypi?%3Aaction=list_classifiers
classifiers = [
    "License :: OSI Approved :: Apache Software License",
    "Operating System :: OS Independent",
    "Programming Language :: Python :: 3 :: Only",
    "Programming Language :: Python :: 3.10",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Programming Language :: Python :: 3.8",
    "Programming Language :: Python :: 3.9",
    "Programming Language :: Python",
]
dependencies = [
    "PyCryptodome",
    "html2text",
    "numpy",
    "openai>=1",
    "pydantic>=2",
    "pypdf",
    "tiktoken>=0.4.0",
]
description = "LLM Chain for answering questions from docs"
dynamic = ["optional-dependencies"]
keywords = ["question answering"]
license = {file = "LICENSE"}
maintainers = [
    {email = "jamesbraza@gmail.com", name = "James Braza"},
    {email = "white.d.andrew@gmail.com", name = "Andrew White"},
]
name = "paper-qa"
readme = "README.md"
requires-python = ">=3.8"
urls = {repository = "https://github.com/whitead/paper-qa"}
version = "4.6.0"

[tool.codespell]
check-filenames = true
check-hidden = true
# SEE: https://github.com/codespell-project/codespell/issues/1212#issuecomment-1744768533
ignore-regex = ".{1024}|.*codespell-ignore.*"
ignore-words-list = "aadd,ser"

[tool.mypy]
# Type-checks the interior of functions without type annotations.
check_untyped_defs = true
# Allows enabling one or multiple error codes globally. Note: This option will
# override disabled error codes from the disable_error_code option.
enable_error_code = [
    "ignore-without-code",
    "mutable-override",
    "redundant-cast",
    "redundant-expr",
    "redundant-self",
    "truthy-bool",
    "truthy-iterable",
    "unreachable",
    "unused-awaitable",
    "unused-ignore",
]
# Shows a short summary line after error messages.
error_summary = false
# Use visually nicer output in error messages: use soft word wrap, show source
# code snippets, and show error location markers.
pretty = true
# Shows column numbers in error messages.
show_column_numbers = true
# Shows error codes in error messages.
# SEE: https://mypy.readthedocs.io/en/stable/error_codes.html#error-codes
show_error_codes = true
# Prefixes each error with the relevant context.
show_error_context = true
# Warns about casting an expression to its inferred type.
warn_redundant_casts = true
# Shows a warning when encountering any code inferred to be unreachable or
# redundant after performing type analysis.
warn_unreachable = true
# Warns about per-module sections in the config file that do not match any
# files processed when invoking mypy.
warn_unused_configs = true
# Warns about unneeded `# type: ignore` comments.
warn_unused_ignores = true

[[tool.mypy.overrides]]
# Suppresses error messages about imports that cannot be resolved.
ignore_missing_imports = true
# Per-module configuration options
module = [
    "fitz",
    "pyzotero",  # SEE: https://github.com/urschrei/pyzotero/issues/110
    "sentence_transformers",  # SEE: https://github.com/UKPLab/sentence-transformers/issues/1723
]

[tool.pytest.ini_options]
# List of directories that should be searched for tests when no specific directories,
# files or test ids are given in the command line when executing pytest from the rootdir
# directory. File system paths may use shell-style wildcards, including the recursive **
# pattern.
testpaths = ["tests"]

[tool.ruff]
# Line length to use when enforcing long-lines violations (like `E501`).
line-length = 120
# Enable application of unsafe fixes.
unsafe-fixes = true

[tool.ruff.lint]
# List of rule codes that are unsupported by Ruff, but should be preserved when
# (e.g.) validating # noqa directives. Useful for retaining # noqa directives
# that cover plugins not yet implemented by Ruff.
ignore = [
    "ANN",  # Don't care to enforce typing
    "BLE001",  # Don't care to enforce blind exception catching
    "COM812",  # Trailing comma with black leads to wasting lines
    "D100",  # D100, D101, D102, D103, D104, D105, D106, D107: don't always need docstrings
    "D101",
    "D102",
    "D103",
    "D104",
    "D105",
    "D106",
    "D107",
    "D203",  # Keep docstring next to the class definition (covered by D211)
    "D212",  # Summary should be on second line (opposite of D213)
    "D402",  # It's nice to reuse the method name
    "D406",  # Google style requires ":" at end
    "D407",  # We aren't using numpy style
    "D413",  # Blank line after last section. -> No blank line
    "DTZ",  # Don't care to have timezone safety
    "EM",  # Overly pedantic
    "ERA001",  # Don't care to prevent commented code
    "FBT001",  # FBT001, FBT002: overly pedantic
    "FBT002",
    "FIX",  # Don't care to prevent TODO, FIXME, etc.
    "FLY002",  # Can be less readable
    "G004",  # f-strings are convenient
    "INP001",  # Can use namespace packages
    "N803",  # Want to use 'N', or 'L',
    "N806",  # Want to use 'N', or 'L',
    "PLR0913",
    "PTH",  # Overly pedantic
    "S311",  # Ok to use python random
    "SLF001",  # Overly pedantic
    "T201",  # Overly pedantic
    "TCH001",  # TCH001, TCH002, TCH003: don't care to enforce type checking blocks
    "TCH002",
    "TCH003",
    "TD002",  # Don't care for TODO author
    "TD003",  # Don't care for TODO links
    "TID252",  # Allow relative imports for packaging
    "TRY003",  # Overly pedantic
]
select = ["ALL"]
unfixable = [
    "B007",  # While debugging, unused loop variables can be useful
    "ERA001",  # While debugging, temporarily commenting code can be useful
    "F401",  # While debugging, unused imports can be useful
    "F841",  # While debugging, unused locals can be useful
]

[tool.ruff.lint.flake8-annotations]
mypy-init-return = true

[tool.ruff.lint.per-file-ignores]
"tests/*.py" = [
    "PLR2004",  # Tests can have magic values
    "S101",  # Tests can have assertions
]

[tool.ruff.lint.pycodestyle]
# The maximum line length to allow for line-length violations within
# documentation (W505), including standalone comments.
max-doc-length = 120  # Match line-length

[tool.ruff.lint.pydocstyle]
# Whether to use Google-style or NumPy-style conventions or the PEP257
# defaults when analyzing docstring sections.
convention = "google"

[tool.setuptools.dynamic.optional-dependencies.dev]
file = ["dev-requirements.txt"]

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

[tool.tomlsort]
all = true
in_place = true
spaces_before_inline_comment = 2  # Match Python PEP 8
spaces_indent_inline_array = 4  # Match Python PEP 8
trailing_comma_inline_array = true