pyproject.toml

[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"

[project]
name = "parallel-web-tools"
version = "0.3.0"
description = "Parallel Tools: CLI and Python SDK for AI-powered web intelligence"
readme = "README.md"
requires-python = ">=3.10"
license = "MIT"
authors = [
    { name = "Parallel", email = "support@parallel.ai" }
]
keywords = ["ai", "data-enrichment", "parallel", "etl", "data-pipeline", "web-search", "llm"]
classifiers = [
    "Development Status :: 4 - Beta",
    "Intended Audience :: Developers",
    "License :: OSI Approved :: MIT License",
    "Operating System :: OS Independent",
    "Programming Language :: Python :: 3",
    "Programming Language :: Python :: 3.10",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Programming Language :: Python :: 3.13",
    "Programming Language :: Python :: 3.14",
    "Topic :: Scientific/Engineering :: Artificial Intelligence",
    "Topic :: Database",
]

dependencies = [
    "parallel-web>=0.6.0",
    "python-dotenv>=1.0.0",
    # CLI dependencies (minimal - search, extract, enrich with CLI args)
    "click>=8.1.0",
    "rich>=13.0.0",
    "httpx>=0.25.0",
]

[project.urls]
Homepage = "https://github.com/parallel-web/parallel-web-tools"
Documentation = "https://docs.parallel.ai"
Repository = "https://github.com/parallel-web/parallel-web-tools"
Issues = "https://github.com/parallel-web/parallel-web-tools/issues"

[project.scripts]
parallel-cli = "parallel_web_tools.cli:main"

[project.optional-dependencies]
# CLI extras: YAML config files and interactive planner
cli = [
    "pyyaml>=6.0.0",
    "questionary>=2.0.0",
]
# Polars integration for data processing
polars = [
    "polars>=1.37.0",
    "pyarrow>=18.0.0",
]
# Pandas integration (for Spark UDFs)
pandas = [
    "pandas>=2.3.0",
]
# DuckDB integration (includes polars for data handling)
duckdb = [
    "parallel-web-tools[cli,polars]",
    "duckdb>=1.0.0",
    "nest-asyncio>=1.6.0",
]
# Snowflake integration
snowflake = [
    "snowflake-connector-python>=3.0.0",
]
# SQLAlchemy-based BigQuery connector (for YAML configs and ETL)
bigquery = [
    "parallel-web-tools[cli]",
    "sqlalchemy>=2.0.0",
    "sqlalchemy-bigquery>=1.11.0",
]
# BigQuery native UDF deployment (Cloud Functions + Remote Functions)
# Note: Requires gcloud CLI to be installed. See docs/bigquery-setup.md
bigquery-native = []  # No Python deps, just CLI tools
# Apache Spark integration
spark = [
    "parallel-web-tools[pandas]",
    "pyspark>=3.4.0",
]
# All features (recommended for full CLI experience)
all = [
    "parallel-web-tools[cli,polars,duckdb,snowflake,bigquery]",
]
# Development
dev = [
    "parallel-web-tools[all,spark]",
    "pytest>=9.0.0",
    "pytest-cov>=7.0.0",
    "pyinstaller>=6.20.0",
    "pre-commit>=4.6.0",
    "ruff>=0.15.0",
    "ty>=0.0.33",
]

[tool.hatch.build.targets.wheel]
packages = ["parallel_web_tools"]
artifacts = [
    "*.sql",
    "*.txt",  # requirements.txt for cloud function
]

[tool.hatch.build.targets.sdist]
include = [
    "/parallel_web_tools",
]

[tool.ty.src]
# Exclude files that have external dependencies not in the main project
exclude = [
    "**/integrations/bigquery/cloud_function/**",  # Separate deployment with flask, functions_framework
    "**/integrations/spark/**",  # PySpark type stubs are incomplete (pandas_udf overloads)
    "**/integrations/duckdb/udf.py",  # DuckDB UDF uses type="arrow" not in type stubs
    "scripts/runtime_hook_ssl.py",  # PyInstaller runtime hook with sys._MEIPASS
    "notebooks/**",  # Jupyter notebooks with Databricks display()
    "examples/**",  # Example scripts
]

# SDK returns TaskRunEvent | ErrorEvent union; code narrows via event.type check
# but ty can't follow string-based narrowing on .type attribute
[[tool.ty.overrides]]
include = ["**/core/batch.py"]
[tool.ty.overrides.rules]
unresolved-attribute = "ignore"

# Polars .collect() returns InProcessQuery | DataFrame in newer polars stubs
[[tool.ty.overrides]]
include = ["**/integrations/polars/**"]
[tool.ty.overrides.rules]
invalid-argument-type = "ignore"

[tool.ruff]
line-length = 120
target-version = "py310"

[tool.ruff.lint]
select = [
    "E",      # pycodestyle errors
    "W",      # pycodestyle warnings
    "F",      # pyflakes
    "I",      # isort
    "B",      # flake8-bugbear
    "UP",     # pyupgrade
]
ignore = [
    "E501",   # line too long (handled by formatter)
    "B008",   # do not perform function calls in argument defaults
]

[tool.ruff.lint.isort]
known-first-party = ["parallel_web_tools"]

[tool.ruff.lint.per-file-ignores]
"examples/*" = ["E402"]  # Allow imports not at top for tutorial scripts

[dependency-groups]
dev = [
    "ipykernel>=7.2.0",
    "pyinstaller>=6.20.0",
    "tach>=0.34.1",
    "ty>=0.0.33",
]