ruff.toml

# Ruff configuration for METPO
# See: https://docs.astral.sh/ruff/configuration/

# Exclude directories
exclude = [
    ".git",
    ".venv",
    "__pycache__",
    "build",
    "dist",
    ".eggs",
    "*.egg-info",
    "src/ontology",  # ODK-generated ontology files
    "notebooks",      # Jupyter notebooks
    "docs",           # Documentation including notebooks and experimental scripts
    "large",          # Reference data scripts
    "downloads",
    "local",
]

# Python version target
target-version = "py311"

# Line length
line-length = 100

[lint]
# Enable rules
select = [
    "E",    # pycodestyle errors
    "W",    # pycodestyle warnings
    "F",    # pyflakes
    "I",    # isort
    "N",    # pep8-naming
    "UP",   # pyupgrade
    "B",    # flake8-bugbear
    "C4",   # flake8-comprehensions
    "DTZ",  # flake8-datetimez
    "T10",  # flake8-debugger
    "EXE",  # flake8-executable
    "ISC",  # flake8-implicit-str-concat
    "ICN",  # flake8-import-conventions
    "PIE",  # flake8-pie
    "PT",   # flake8-pytest-style
    "Q",    # flake8-quotes
    "RSE",  # flake8-raise
    "RET",  # flake8-return
    "SIM",  # flake8-simplify
    "PTH",  # flake8-use-pathlib
    "PL",   # pylint
    "RUF",  # ruff-specific rules
]

# Ignore specific rules
ignore = [
    "E501",   # line-too-long (handled by formatter)
    "PLR0913", # too-many-arguments
    "PLR2004", # magic-value-comparison
    "PT009",   # pytest-unittest-assertion (we use assert)
    "RET504",  # unnecessary-assign (sometimes clearer)
]

# Allow autofix for all enabled rules
fixable = ["ALL"]
unfixable = []

# Allow unused variables when underscore-prefixed
dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"

[format]
# Use double quotes for strings
quote-style = "double"

# Indent with spaces
indent-style = "space"

# Like Black, respect magic trailing commas
skip-magic-trailing-comma = false

# Like Black, automatically detect the appropriate line ending
line-ending = "auto"

[lint.per-file-ignores]
# Ignore import errors in __init__.py
"__init__.py" = ["F401"]
# Allow print statements in scripts
"metpo/tools/*.py" = ["T201"]
"metpo/scripts/*.py" = ["T201"]
"literature_mining/scripts/*.py" = ["T201"]
# Complexity warnings for specific analysis/utility scripts - refactoring would require significant restructuring
"metpo/bactotraits/bactotraits_metpo_set_difference.py" = ["PLR0915", "PLR0912"]
"metpo/bactotraits/create_bactotraits_file_versions.py" = ["PLR0915"]
"metpo/analysis/assess_ontology_by_api_search.py" = ["PLR0915"]
"metpo/analysis/extract_definitions_from_mappings.py" = ["PLR0915"]
"metpo/analysis/analyze_ontology_value.py" = ["PLR0915", "PLR0912"]
# Ignore complexity checks in analysis scripts
"metpo/analysis/*.py" = ["PLR0912", "PLR0915"]
"literature_mining/analyze_*.py" = ["PLR0912", "PLR0915"]
# Ignore complexity and minor issues in bactotraits data processing scripts
"metpo/bactotraits/*.py" = ["PLR0911", "PLR0912", "PLR0915", "SIM102", "B007", "RUF001", "N806", "EXE002"]
# Database scripts - complexity and minor issues
"metpo/database/audit_chromadb.py" = ["PLR0912", "PLR0915"]
"metpo/database/combine_chromadb.py" = ["PLC0415", "PTH110"]
"metpo/database/filter_ols_chromadb.py" = ["PLR0912", "PLR0915"]
"metpo/database/migrate_to_chromadb_resilient.py" = ["PLR0912", "PLR0915", "SIM115"]
# Literature mining analysis - complexity and style issues
"metpo/literature_mining/analysis/analyze_coverage_by_source_type.py" = ["PLR0915"]
"metpo/literature_mining/analysis/analyze_extractions.py" = ["PLR0912"]
"metpo/literature_mining/analysis/analyze_metpo_database_alignment.py" = ["PLR0915"]
"metpo/literature_mining/analysis/analyze_metpo_efficiency.py" = ["PLR0912", "PLR0915", "SIM102"]
"metpo/literature_mining/analysis/analyze_metpo_grounding.py" = ["PLR0912", "PLR0915", "SIM102"]
"metpo/literature_mining/analysis/analyze_metpo_grounding_filtered.py" = ["PLR0912", "PLR0915"]
"metpo/literature_mining/analysis/compare_extractions.py" = ["PLR0915"]
"metpo/literature_mining/analysis/metpo_assessor.py" = ["B007", "PLR0912", "PLR0915"]
"metpo/literature_mining/analysis/validate_extractions.py" = ["SIM102"]
"metpo/literature_mining/analysis/visualize_ner.py" = ["PLC0415", "PLR0911", "PLR0912"]
# Literature mining scripts
"metpo/literature_mining/scripts/count_extraction_results.py" = ["PLC0415"]
"metpo/literature_mining/scripts/dedupe_and_filter_abstracts.py" = ["PLR0912"]
"metpo/literature_mining/scripts/dedupe_by_content.py" = ["B007", "PLR0912"]
"metpo/literature_mining/scripts/fetch_abstracts_from_dois.py" = ["PLC0415"]
# Pipeline scripts - complexity and minor issues
"metpo/pipeline/analyze_coherence_results.py" = ["PLR0915"]
"metpo/pipeline/analyze_sibling_coherence.py" = ["PLR0912", "PLR0915", "PTH110", "RUF012", "SIM108"]
"metpo/pipeline/categorize_ontologies.py" = ["PLC0206", "PLR0912", "PLR0915"]
"metpo/pipeline/chromadb_semantic_mapper.py" = ["PLC0415", "PLR0912", "PLR0915", "RUF001"]
"metpo/pipeline/embed_ontology_to_chromadb.py" = ["PTH119"]
"metpo/pipeline/fetch_ontology_names.py" = ["SIM113"]
# Presentation scripts
"metpo/presentations/analyze_kg_microbe_metpo.py" = ["PLR0915"]
"metpo/presentations/analyze_ontology_landscape.py" = ["PLR0915"]
"metpo/presentations/calculate_minimum_import_set.py" = ["B007", "PLR0912", "PLR0915"]
# General scripts - complexity warnings
"metpo/scripts/bootstrap_definition_enrichment.py" = ["PLR0912", "PLR0915"]
"metpo/scripts/compare_definitions_with_hierarchy.py" = ["PLR0912", "PLR0915"]
"metpo/scripts/definition_work/fetch_source_metadata.py" = ["PLR0912", "PLR0915"]
"metpo/scripts/definition_work/merge_sparql_with_metadata.py" = ["SIM117"]
"metpo/scripts/find_best_definitions.py" = ["PLR0912", "PLR0915"]
"metpo/scripts/find_best_definitions_comprehensive.py" = ["PLR0912", "PLR0915"]
"metpo/scripts/iterative_definition_improvement.py" = ["PLR0912", "PLR0915", "SIM102"]
"metpo/scripts/plot_embedding_similarity.py" = ["PLR0915"]
"metpo/scripts/propose_definitions_with_llm.py" = ["PLR0915"]
"metpo/scripts/qc_metpo_sheets.py" = ["PLR0915", "RUF001"]
# Tools
"metpo/tools/import_bactotraits.py" = ["PLR0912", "PLR0915"]

[lint.isort]
known-first-party = ["metpo"]
force-single-line = false
force-sort-within-sections = false