pyproject.toml

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

[tool.codespell]
check-filenames = true
check-hidden = true
ignore-words-list = "astroid,gallary,momento,narl,ot,rouge"
# Feel free to un-skip examples, and experimental, you will just need to
# work through many typos (--write-changes and --interactive will help)
skip = "./examples,./experimental,*.csv,*.html,*.json,*.jsonl,*.pdf,*.txt,*.ipynb"

[tool.mypy]
disallow_untyped_defs = true
# Remove venv skip when integrated with pre-commit
exclude = ["build", "examples", "notebooks", "venv"]
ignore_missing_imports = true

[tool.poetry]
authors = ["Jerry Liu <jerry@llamaindex.ai>"]
classifiers = [
    "Topic :: Scientific/Engineering :: Artificial Intelligence",
    "Topic :: Software Development :: Libraries :: Application Frameworks",
    "Topic :: Software Development :: Libraries :: Python Modules",
]
description = "Interface between LLMs and your data"
documentation = "https://docs.llamaindex.ai/en/stable/"
homepage = "https://llamaindex.ai"
keywords = ["LLM", "NLP", "RAG", "data", "devtools", "index", "retrieval"]
license = "MIT"
maintainers = [
    "Jerry Liu <jerry@llamaindex.ai>",
    "Logan Markewich <logan@llamaindex.ai>",
    "Simon Suo <simon@llamaindex.ai>",
    "Sourabh Desai <sourabh@llamaindex.ai>",
]
name = "llama-index"
packages = [{include = "llama_index"}]
readme = "README.md"
repository = "https://github.com/run-llama/llama_index"
version = "0.8.47"

[tool.poetry.dependencies]
SQLAlchemy = {extras = ["asyncio"], version = ">=1.4.49"}
dataclasses-json = "^0.5.7"
fsspec = ">=2023.5.0"
langchain = ">=0.0.303"
nest-asyncio = "^1.5.8"
nltk = "^3.8.1"
numpy = "*"
openai = ">=0.26.4"
pandas = "*"
python = ">=3.8.1,<3.12"
tenacity = ">=8.2.0,<9.0.0"
tiktoken = ">=0.3.3"
typing-extensions = ">=4.5.0"
typing-inspect = ">=0.8.0"
urllib3 = "<2"

[tool.poetry.extras]
dataloaders = [
    "EbookLib",
    "Pillow",
    "bark",
    "beautifulsoup4",
    "discord.py",
    "elevenlabs",
    "feedparser",
    "google-api-python-client",
    "google-auth-httplib2",
    "google-auth-oauthlib",
    "html2text",
    "mailbox",
    "nbconvert",
    "psychicapi",
    "pydub",
    "pypdf",
    "python-pptx",
    "slack-sdk",
    "trafilatura",
    "tree-sitter-languages",
    "tweepy",
    "wikipedia",
    "youtube-transcript-api",
]
local_models = [
    "optimum",
    "sentencepiece",
    "transformers",
]
postgres = [
    "asyncpg",
    "pgvector",
    "psycopg-binary",
]
query_tools = [
    "guardrails-ai",
    "guidance",
    "jsonpath-ng",
    "rank-bm25",
    "scikit-learn",
    "spacy",
]

[tool.poetry.group.callbacks]
optional = true

[tool.poetry.group.callbacks.dependencies]
aim = {optional = true, version = "^4.0.3"}
arize-phoenix = {optional = true, version = "^0.0.44"}
wandb = {optional = true, version = "^0.15.12"}

[tool.poetry.group.dataloaders]
optional = true

[tool.poetry.group.dataloaders.dependencies]
EbookLib = {optional = true, version = "^0.18"}
Pillow = {optional = true, version = "<=9.5.0"}
bark = {optional = true, version = "^0.1.5"}
beautifulsoup4 = {optional = true, version = "^4.12.2"}
"discord.py" = {optional = true, version = "^2.3.2"}
elevenlabs = {optional = true, version = "^0.2.24"}
feedparser = {optional = true, version = "^6.0.10"}
google-api-python-client = {optional = true, version = "^2.102.0"}
google-auth-httplib2 = {optional = true, version = "^0.1.1"}
google-auth-oauthlib = {optional = true, version = "^1.1.0"}
# steamship = {version = "^2.17.31", optional = true}  # conflicts with weaviate-client
html2text = {optional = true, version = "^2020.1.16"}
mailbox = {optional = true, version = "^0.4"}
nbconvert = {optional = true, version = "^6.5.0"}
psychicapi = {optional = true, version = "^0.8.4"}
pydub = {optional = true, version = "^0.25.1"}
pypdf = {optional = true, version = "^3.16.3"}
python-pptx = {optional = true, version = "^0.6.22"}
slack-sdk = {optional = true, version = "^3.23.0"}
trafilatura = {optional = true, version = "^1.6.2"}
# openai-whisper = {git = "https://github.com/openai/whisper.git", optional = true}  # conflicts with litellm due to tiktoken
tree-sitter-languages = {optional = true, version = "^1.7.0"}
tweepy = {optional = true, version = "^4.14.0"}
wikipedia = {optional = true, version = "^1.4.0"}
youtube-transcript-api = {optional = true, version = "^0.6.1"}

[tool.poetry.group.dev.dependencies]
beautifulsoup4 = "^4.12.2"  # needed for tests
black = {extras = ["jupyter"], version = "<=23.9.1,>=23.7.0"}
codespell = {extras = ["toml"], version = ">=v2.2.6"}
google-generativeai = {python = ">=3.9,<3.12", version = "^0.2.1"}
ipython = "8.10.0"
mypy = "0.991"
pre-commit = "3.2.0"
pylint = "2.15.10"
pymongo = "^4.5.0"  # needed for tests
pytest = "7.2.1"
pytest-asyncio = "0.21.0"
pytest-dotenv = "0.5.2"
pytest-mock = "3.11.1"
rake-nltk = "1.0.6"
ruff = "0.0.292"
types-PyYAML = "^6.0.12.12"
types-redis = "4.5.5.0"
types-requests = "2.28.11.8"
types-setuptools = "67.1.0.0"
vellum-ai = "^0.0.42"

[tool.poetry.group.docs]
optional = true

[tool.poetry.group.docs.dependencies]
autodoc-pydantic = "<=1.9.0"
docutils = "<0.17"
furo = ">=2023.3.27"
m2r2 = "0.3.2"
myst-nb = "0.17.2"
myst-parser = "0.18.1"
pydantic = "<2.0.0"
sphinx = ">=4.3.0"
sphinx-autobuild = "^2021.3.14"

[tool.poetry.group.embeddings]
optional = true

[tool.poetry.group.embeddings.dependencies]
# clarifai = {version = "^9.9.0", optional = true}  # conflicts with chromadb
InstructorEmbedding = {optional = true, version = "^1.0.1"}
sentence-transformers = {optional = true, version = "^2.2.2"}

[tool.poetry.group.llms]
optional = true

[tool.poetry.group.llms.dependencies]
anthropic = {optional = true, version = "^0.3.11"}
# clarifai = {version = "^9.9.0", optional = true}  # conflicts with chromadb
cohere = {optional = true, version = "^4.27"}
google-generativeai = {optional = true, python = ">=3.9,<3.12", version = "^0.2.1"}
gradientai = {optional = true, python = ">=3.10,<4.0", version = "^1.1.0"}
litellm = {optional = true, version = "^0.5.6"}
llama-cpp-python = {optional = true, version = "^0.2.11"}
llamaapi = {optional = true, python = ">=3.9,<3.12", version = "^0.1.36"}
monsterapi = {optional = true, version = "^1.0.2"}
portkey-ai = {optional = true, version = "^0.1.52"}
predibase = {optional = true, version = "^2023.10.3"}
replicate = {optional = true, version = "^0.15.4"}
sseclient-py = {optional = true, version = "^1.8.0"}
vellum-ai = {optional = true, version = "^0.0.42"}
xinference = {optional = true, version = "^0.5.2"}

[tool.poetry.group.metadata_extractors]
optional = true

[tool.poetry.group.metadata_extractors.dependencies]
marvin = {optional = true, python = ">=3.9,<3.12", version = "^1.5.1"}
span-marker = {optional = true, version = "^1.4.0"}

[tool.poetry.group.model_utils]
optional = true

[tool.poetry.group.model_utils.dependencies]
optimum = {extras = ["onnxruntime"], optional = true, version = "^1.13.2"}
sentencepiece = {optional = true, version = "^0.1.99"}
transformers = {extras = ["torch"], optional = true, version = "^4.34.0"}

[tool.poetry.group.query]
optional = true

[tool.poetry.group.query.dependencies]
# guardrails-ai = {version = "^0.2.4", optional = true}  # conflicts with bageldb due to pydantic
guidance = {optional = true, version = "^0.0.64"}
jsonpath-ng = {optional = true, version = "^1.6.0"}
rank-bm25 = {optional = true, version = "^0.2.2"}
scikit-learn = {optional = true, version = "<1.3.0"}
spacy = {optional = true, version = "^3.7.1"}

[tool.poetry.group.storage]
optional = true

[tool.poetry.group.storage.dependencies]
boto3 = {optional = true, version = ">=1.28.0"}
google-cloud-firestore = {optional = true, version = "^2.12.0"}
kuzu = {optional = true, version = "^0.0.9"}
nebula3-python = {optional = true, version = "^3.4.0"}
neo4j = {optional = true, version = "^5.13.0"}
networkx = {optional = true, version = "^3.1"}
pymongo = {optional = true, version = "^4.5.0"}
redis = {optional = true, version = ">=4.4.4"}

[tool.poetry.group.vector_stores]
optional = true

[tool.poetry.group.vector_stores.dependencies]
asyncpg = {optional = true, version = "^0.28.0"}
awadb = {optional = true, version = "^0.3.10"}
azure-search-documents = {optional = true, version = "^11.3.0"}
betabageldb = {optional = true, version = "^0.2.34"}
cassio = {optional = true, version = "^0.1.3"}
chromadb = {optional = true, version = ">=0.4.0"}
clickhouse-connect = {optional = true, version = "^0.6.14"}
deeplake = {optional = true, version = "^3.7.3"}
docarray = {optional = true, version = "^0.39.0"}
elasticsearch = {optional = true, version = "^8.10.0"}
faiss = {optional = true, version = "^1.5.3"}
metal-sdk = {optional = true, version = "^2.2.2"}
opensearch-py = {optional = true, version = "^2.3.1"}
pgvector = {optional = true, version = "^0.1.0"}
pinecone-client = {optional = true, version = "^2.2.4"}
psycopg-binary = {optional = true, version = "^3.1.12"}
pyepsilla = {optional = true, version = "^0.1.12"}
pymilvus = {optional = true, version = "^2.3.1"}
qdrant-client = {optional = true, version = "^1.6.0"}
rockset = {optional = true, version = "^2.1.0"}
tair = {optional = true, version = "^1.3.7"}
timescale-vector = {optional = true, version = "^0.0.1"}
typesense = {optional = true, version = "^0.17.0"}
vecs = {optional = true, version = "^0.4.1"}
weaviate-client = {optional = true, version = "^3.24.2"}
zep-python = {optional = true, version = "^1.2.1"}

[tool.ruff]
exclude = [
    "examples",
    "notebooks",
]
ignore = [
    "COM812",  # Too aggressive
    "D212",  # Using D213
    "D417",  # Too aggressive
    "F541",  # Messes with prompts.py
    "TCH002",
    "UP006",  # Messes with pydantic
]
# Feel free to add more here
select = [
    "ANN204",
    "B009",
    "B010",
    "B011",
    "B013",
    "B014",
    "C4",
    "COM812",
    "COM819",
    "D201",
    "D202",
    "D203",
    "D204",
    "D207",
    "D208",
    "D209",
    "D211",
    "D213",
    "D214",
    "D215",
    "D3",
    "D4",
    "E7",
    "EXE004",
    "F504",
    "F541",
    "F632",
    "FLY",
    "G010",
    "I",
    "PERF1",
    "PIE790",
    "PIE794",
    "PIE808",
    "PIE810",
    "PLC0414",
    "PLE2510",
    "PLE2512",
    "PLE2513",
    "PLE2514",
    "PLE2515",
    "PLR1701",
    "PLR1711",
    "PT001",
    "PT003",
    "PT006",
    "PT02",
    "PTH201",
    "PYI",
    "Q",
    "RET501",
    "RET502",
    "RET503",
    "RET504",
    "RSE",
    "RUF005",
    "RUF010",
    "RUF015",
    "RUF1",
    "SIM101",
    "SIM103",
    "SIM109",
    "SIM118",
    "SIM2",
    "SIM300",
    "SIM9",
    "TCH005",
    "TD006",
    "TID",
    "TRY201",
    "UP",
    "W",
]
target-version = "py38"
unfixable = [
    "ERA001",
]

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

[tool.ruff.pydocstyle]
convention = "google"

[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