pyproject.toml

# NOTE: You have to use single-quoted strings in TOML for regular expressions.
# It's the equivalent of r-strings in Python.  Multiline strings are treated as
# verbose regular expressions by Black.  Use [ ] to denote a significant space
# character.

[project]
name = "openlibrary"
version = "1.0.0"
requires-python = ">=3.12.2,<3.12.3"

[tool.black]
skip-string-normalization = true
target-version = ["py311"]

[tool.codespell]
ignore-words-list = "beng,curren,datas,furst,nd,nin,ot,ser,spects,te,tha,ue,upto,thirdparty"
skip = "./.*,*/ocm00400866,*/read_toc.py,*.it,*.js,*.json,*.mrc,*.page,*.pg_dump,*.po,*.txt,*.xml,*.yml"

[tool.mypy]
ignore_missing_imports = true
pretty = true
scripts_are_modules = true
show_error_codes = true
show_error_context = true
exclude = ["vendor/"]

[[tool.mypy.overrides]]
module = [
  "infogami.*",
  "openlibrary.plugins.worksearch.code"
]
ignore_errors = true

[[tool.mypy.overrides]]
# In general, we want to have typed definitions so we are slowly turning it on for a few files.
module = ["openlibrary.core.wikidata", "openlibrary.core.yearly_reading_goals"]
disallow_untyped_defs = true

[tool.pytest.ini_options]
strict = true
asyncio_mode = "strict"
asyncio_default_fixture_loop_scope = "function"
filterwarnings = [
    # Ignore warnings from webob, which doesn't have a fix.
    # It comes from webtest, which we will remove when fastapi is done
    'ignore::DeprecationWarning:webob',
    # Ignore sqlite3 default timestamp converter warnings from web.py (Python 3.12 deprecation)
    'ignore:The default timestamp converter is deprecated:DeprecationWarning:web',
]
addopts = "-m 'not integration'"
markers = [
    "integration: marks tests as integration tests that require external services (deselect with '-m \"not integration\"')"
]

[tool.ruff]
target-version = "py312"
extend-exclude = [
  "./.*",
  "vendor",
]
lint.ignore = [
  "B007",    # unused-loop-control-variable - not a big deal for us
  "B023",    # function-uses-loop-variable - doesn't matter for our lambdas but it's a good rule
  "B904",    # raise-without-from-inside-except - not sure but it seems fine for now
  "PERF401", # manual-list-comprehension - we prefer readability
  "PIE790",  # unnecessary-placeholder - could be removed but not much gain
  "PLW0603", # global-statement - we use it a lot...
  "PLW2901", # redefined-loop-name - would be a lot of work to fix
  "RUF005",  # collection-literal-concatenation - this works for us as is
  "SIM108",  # if-else-block-instead-of-if-exp - we prefer the readability instead of ternaries
  "SLF001",  # private-member-access - unfortunately we use these quite often for infogami
  "UP031",   # printf-string-formatting - a ton of work to fix to 224 errors
  # TODO: these could be fixed one day
  "TC003",   # Move standard library import into a type-checking block. Should enable later
  "TC006",   # runtime-cast-value - we don't do this now but could in a future PR
  "B905",    # zip-without-explicit-strict - TODO: seems like we should fix this one
  "E402",    # module-import-not-at-top-of-file - TODO: we should probably enable this and and add exceptions
    # openlibrary/plugins/admin/code.py - maybe it can be fixed
    # openlibrary/plugins/openlibrary/code.py - unclear but has many
    # openlibrary/plugins/openlibrary/filters.py - can fix
    # openlibrary/plugins/upstream/jsdef.py - can remove
    # openlibrary/plugins/upstream/utils.py - unclear but it's huge
    # scripts/tests/test_affiliate_server.py - probably needed for side effects
    # scripts/tests/test_solr_updater.p - probably needed for side effects
  "PLC0415", # import-outside-top-level we really need to turn this on but it's a decently big change
]
line-length = 162
lint.select = [
  "ASYNC",   # flake8-async
  "B",       # flake8-bugbear
  "BLE",     # flake8-blind-except
  "C4",      # flake8-comprehensions
  "C90",     # McCabe cyclomatic complexity
  "E",       # pycodestyle
  "F",       # Pyflakes
  "FA",      # flake8-future-annotations
  "FAST",    # FastAPI rules
  "FLY",     # flynt
  "FURB",    # refurb
  "G010",    # flake8-logging-format
  "I",       # isort
  "ICN",     # flake8-import-conventions
  "INT",     # flake8-gettext
  "ISC",     # flake8-implicit-str-concat
  "PERF",    # Perflint
  "PIE",     # flake8-pie
  "PL",      # Pylint
  "PT",      # flake8-pytest-style
  "PYI",     # flake8-pyi
  "RSE",     # flake8-raise
  "RUF",     # Ruff-specific rules
  "SIM",     # flake8-simplify
  "SLF",     # flake8-self
  "SLOT",    # flake8-slots
  "T10",     # flake8-debugger
  "TC",      # flake8-type-checking
  "UP",      # pyupgrade
  "W",       # pycodestyle
  "YTT",     # flake8-2020
  # "A",     # flake8-builtins
  # "AIR",   # Airflow
  # "ANN",   # flake8-annotations
  # "ARG",   # flake8-unused-arguments
  # "COM",   # flake8-commas
  # "CPY",   # Copyright-related rules
  # "D",     # pydocstyle
  # "DJ",    # flake8-django
  # "DTZ",   # flake8-datetimez
  # "EM",    # flake8-errmsg
  # "ERA",   # eradicate
  # "EXE",   # flake8-executable
  # "FBT",   # flake8-boolean-trap
  # "FIX",   # flake8-fixme
  # "G",     # flake8-logging-format
  # "INP",   # flake8-no-pep420
  # "N",     # pep8-naming
  # "NPY",   # NumPy-specific rules
  # "PD",    # pandas-vet
  # "PGH",   # pygrep-hooks
  # "PTH",   # flake8-use-pathlib
  # "Q",     # flake8-quotes
  # "RET",   # flake8-return
  # "S",     # flake8-bandit
  # "T20",   # flake8-print
  # "TD",    # flake8-todos
  # "TID",   # flake8-tidy-imports
  # "TRY",   # tryceratops
]

[tool.ruff.lint.mccabe]
max-complexity = 28


[tool.ruff.lint.pylint]
allow-magic-value-types = ["bytes", "float", "int", "str"]
max-args = 15
max-branches = 23
max-returns = 14
max-statements = 70

[tool.ruff.lint.per-file-ignores]
"openlibrary/admin/stats.py" = ["BLE001"]
"openlibrary/catalog/add_book/tests/test_add_book.py" = ["PT007"]
"openlibrary/catalog/get_ia.py" = ["BLE001", "E722"]
"openlibrary/catalog/utils/edit.py" = ["E722"]
"openlibrary/catalog/utils/query.py" = ["E722"]
"openlibrary/core/booknotes.py" = ["E722"]
"openlibrary/core/bookshelves.py" = ["BLE001"]
"openlibrary/core/db.py" = ["SIM105"]
"openlibrary/core/helpers.py" = ["BLE001"]
"openlibrary/core/observations.py" = ["PYI024"]
"openlibrary/core/processors/invalidation.py" = ["BLE001"]
"openlibrary/core/ratings.py" = ["E722"]
"openlibrary/core/sponsorships.py" = ["E722"]
"openlibrary/core/stats.py" = ["BLE001"]
"openlibrary/core/vendors.py" = ["B009"]
"openlibrary/coverstore/code.py" = ["E722"]
"openlibrary/i18n/__init__.py" = ["BLE001"]
"openlibrary/plugins/admin/code.py" = ["E722"]
"openlibrary/plugins/admin/mem.py" = ["E722"]
"openlibrary/plugins/admin/memory.py" = ["E722"]
"openlibrary/plugins/admin/services.py" = ["BLE001"]
"openlibrary/plugins/books/dynlinks.py" = ["E722"]
"openlibrary/plugins/books/readlinks.py" = ["E722"]
"openlibrary/plugins/importapi/code.py" = ["BLE001"]
"openlibrary/plugins/ol_infobase.py" = ["BLE001"]
"openlibrary/plugins/openlibrary/code.py" = ["BLE001", "E722"]
"openlibrary/plugins/openlibrary/connection.py" = ["E722"]
"openlibrary/plugins/openlibrary/stats.py" = ["BLE001"]
"openlibrary/plugins/upstream/account.py" = ["BLE001"]
"openlibrary/plugins/upstream/borrow.py" = ["BLE001", "E722"]
"openlibrary/plugins/upstream/models.py" = ["BLE001"]
"openlibrary/plugins/upstream/utils.py" = ["BLE001"]
"openlibrary/solr/solr_types.py" = ["UP007"]
"openlibrary/utils/retry.py" = ["BLE001"]
"openlibrary/utils/open_syllabus_project.py" = ["BLE001"]
"openlibrary/utils/schema.py" = ["PERF402"]
"openlibrary/utils/tests/test_retry.py" = ["PT012", "PT017"]
"scripts/affiliate_server*.py" = ["SIM105"]
"scripts/copydocs.py" = ["BLE001", "PYI024"]
"scripts/delete_import_items.py" = ["BLE001"]
"scripts/import_book_covers.py" = ["BLE001"]
"scripts/lc_marc_update.py" = ["E722"]
"scripts/manage-imports.py" = ["BLE001"]
"scripts/sitemaps/sitemap.py" = ["BLE001"]
"scripts/solr_builder/solr_builder/solr_builder.py" = ["PYI024", "PLR0913"]
"scripts/tests/test_obfi.py" = ["E501"]
"tests/*" = ["S101"]

[tool.sqlfluff.core]
dialect = "postgres"
# ── Rule Exclusions ──────────────────────────────────────────────────
# Only rules that conflict with the existing codebase style are excluded.
# Each exclusion includes a rationale so reviewers can evaluate them.
#
# Layout:
#   LT01 - trailing whitespace: already enforced by pre-commit trailing-whitespace hook
#   LT02 - newline at end of file: already enforced by pre-commit end-of-file-fixer hook
#   LT05 - line length (80 chars): existing SQL files (schema, dump scripts) have many lines
#          exceeding 80 chars; enforcing would require major reformatting of stable schema files
# Capitalisation:
#   CP01-CP05 - keyword/function/datatype casing: codebase consistently uses lowercase
#               keywords throughout; enforcing upper-case would touch every SQL file
# Aliasing:
#   AL01 - implicit aliasing: codebase omits explicit "AS" for table aliases, valid Postgres style
#   AL02 - implicit column aliasing: same as AL01 but for columns
#   AL03 - self-alias (e.g. "x AS x"): existing dump scripts use self-aliases for clarity
#   AL09 - self-aliased columns: same pattern as AL03, catches additional self-alias variants
# References:
#   RF04 - keywords as identifiers: columns named "key", "type", "action", "created" are
#          deeply embedded in the schema and cannot be renamed
# Structure:
#   ST06 - SELECT column ordering: dump scripts' column order matches the public dump API
#          documented at openlibrary.org/developers/dumps
# Ambiguous:
#   AM05 - fully qualified joins: some dump scripts use implicit join conditions; these are
#          stable scripts where changing join syntax risks introducing bugs
# Joins:
#   JJ01 - explicit join type: bare JOIN (implicit INNER) is the codebase convention
exclude_rules = "LT01,LT02,LT05,CP01,CP02,CP03,CP04,CP05,AL01,AL02,AL03,AL09,RF04,ST06,AM05,JJ01"

[tool.sqlfluff.indentation]
tab_space_size = 4