pyproject.toml

[tool.poetry]
name = "reactome-chatbot"
version = "0.1.0"
description = "Chatbot for reactome's data portal"
authors = ["Helia Mohammadi <hmohammadi@oicr.on.ca>",
           "Adam Wright <awright@oicr.on.ca>",
           "Greg Hogue <ghogue@oicr.on.ca>"]
license = "Apache2"
readme = "README.md"
packages = [
    { include = "src", from = "." }
]

#lark is needed for langchain -> SelfQueryRetriever
[tool.poetry.dependencies]
python = ">=3.12, <4"
langchain = "^0.3.4"
openai = "^1.12.0"
chromadb = "<1.0"
pandas = "^2.2.1"
pyasn1 = "^0.5.1"
langchain-openai = "^0.2.3"
neo4j = "4.3.6"
python-dotenv = "^1.0.1"
pyfiglet = "^1.0.2"
chainlit = "^2.0.3"
asyncpg = "^0.30.0"
sqlalchemy = "^2.0.30"
langchain-community = "^0.3.3"
langchain-core = "^0.3.13"
langchain-huggingface = "^0.1.0"
einops = "^0.8.0"
boto3 = "^1.34.148"
torch = [
    { version = "2.4.1+cpu", markers = "sys_platform == 'linux' and platform_machine == 'x86_64'", source = "pytorch_cpu" },
    { version = "2.2.*", markers = "sys_platform != 'linux' or platform_machine != 'x86_64'", source = "PyPI" },
]
langchain-chroma = "0.2.3"
langchain-ollama = "^0.2.0"
lark = "^1.2.2"
langgraph = "^0.2.39"
langgraph-checkpoint-postgres = "^2.0.2"
rank-bm25 = "^0.2.2"
psycopg = {extras = ["binary"], version = "^3.2.3"}
pydantic = "^2.10.5"
pyyaml = "^6.0.2"
tavily-python = "^0.5.0"
openpyxl = "^3.1.5"
nltk = "^3.9.1"
beautifulsoup4 = "^4.12.0"
lxml = "^5.0.0"
requests = "^2.32.0"

[tool.poetry.group.dev.dependencies]
ruff = "^0.7.1"
pytest = "^8.3.3"
mypy = "^1.13.0"
black = "^24.10.0"
isort = "^5.13.2"
pandas-stubs = "^2.2.3.241009"
types-requests = "^2.32.0.20241016"
types-pyyaml = "^6.0.12.20241230"
datasets = "^3.2.0"
ragas = "^0.2.11"

[[tool.poetry.source]]
name = "PyPI"
priority = "primary"

[[tool.poetry.source]]
name = "pytorch_cpu"
url = "https://download.pytorch.org/whl/cpu"
priority = "explicit"

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

[tool.ruff]
line-length = 88
target-version = "py312"
src = ["src", "bin"]
extend-exclude = ["embeddings", "records", "data", ".venv"]

[tool.ruff.lint]
# One rule set for the whole repo. E501 is the formatter's job -- it cannot split
# long string literals anyway -- and RUF001 fires on intentional typography inside
# LLM prompt text.
select = [
    "E", "F", "I", "UP", "B", "SIM", "RUF",  # the core set
    "S",      # bandit: security
    "C4",     # comprehensions
    "RET",    # return-statement hygiene
    "PIE", "PT", "ISC", "INT", "ICN", "TID", "A", "LOG", "G", "ERA", "N",
]
ignore = ["E501", "RUF001", "ISC001"]

# Not yet enforced. Each needs a decision rather than a mechanical fix, so they
# are listed here instead of being silently dropped:
#   PTH  ~41 os.path -> pathlib rewrites; mechanical but touches files the
#        retriever work will rewrite. Worth its own commit afterwards.
#   DTZ  datetime.now() without tzinfo. The rate limiter and static-message
#        throttle deliberately store naive local timestamps in user metadata;
#        changing that is a storage-format decision, not a lint fix.
#   ARG  unused arguments -- several are required by LangGraph node signatures.
#   BLE  blind `except Exception` in top-level scripts.
#   C901 two functions over the complexity threshold.

[tool.ruff.lint.isort]
known-first-party = [
    "agent",
    "data_generation",
    "evaluation",
    "retrievers",
    "tools",
    "util",
]

[tool.ruff.lint.per-file-ignores]
# Entry-point scripts intentionally configure the app at import time.
"bin/*" = ["E402"]
# assert is how pytest asserts.
"tests/*" = ["S101"]

[tool.ruff.format]
docstring-code-format = true

[tool.pytest.ini_options]
testpaths = ["tests"]
pythonpath = ["src", "bin"]
addopts = "-q --strict-markers --strict-config"
markers = [
    "requires_embeddings: needs an installed embeddings bundle (see ./bin/embeddings_manager)",
    "requires_retrieval_stack: needs the full langchain/chromadb dependency set installed",
]

[tool.mypy]
python_version = "3.12"
files = ["bin", "src", "tests", ".github", "export_csvs.py"]
exclude = "^(data|embeddings|records)/"
mypy_path = "src"
explicit_package_bases = true
ignore_missing_imports = true
# --- ratchet: currently-enforced floor (the code already largely satisfies this) ---
check_untyped_defs = true
no_implicit_optional = true
strict_equality = true
warn_redundant_casts = true
# Annotation coverage reached 100%, so these are enforced repo-wide rather than
# per-module. There is now one standard for every file.
disallow_untyped_defs = true
disallow_incomplete_defs = true
warn_no_return = true
warn_return_any = true
extra_checks = true
# disallow_untyped_decorators stays off: chainlit's @cl.* decorators are
# untyped upstream, so it only ever fires on bin/chat-chainlit.py.

# --- mypy baseline -------------------------------------------------------------
# Modules that do not yet meet the floor above. Each entry is debt: fix the module,
# then delete its block. Do not add to this list without a TODO.
[[tool.mypy.overrides]]
# TODO(phase-2): every node returns a partial state; the TypedDict needs
# `total=False` or per-node Partial* types.
module = ["agent.profiles.cross_database"]
disable_error_code = ["typeddict-item", "no-any-return"]

[[tool.mypy.overrides]]
# A module may appear in only one override section, so both TODOs live here.
# TODO(phase-2): partial-state returns, as for cross_database above.
# TODO(phase-2): preprocess() narrows BaseState to ReactToMeState, which is an LSP
# violation -- parameters may widen, not narrow. The fix is to make BaseGraphBuilder
# generic in its state type (BaseGraphBuilder[StateT]) so each profile declares its
# own. Deferred because base.py is also modified by the safetycheck and analysis
# branches, and a generics refactor there would collide with both.
module = ["agent.profiles.react_to_me"]
disable_error_code = ["typeddict-item", "override"]

[[tool.mypy.overrides]]
# TODO(phase-2): `EmbeddingEnvironment.get_dir()` returns `Path | None` but the
# parameter is typed `Path`, so an uninstalled bundle passes None straight through.
# Fixed properly by moving the lookup out of the default argument (see ruff B008).
module = [
    "retrievers.reactome.rag",
    "retrievers.uniprot.rag",
    "retrievers.plantreactome.rag",
    "retrievers.userguide.rag",
]
disable_error_code = ["assignment"]