expletives 0.1.0

Curated profanity lists with provenance, categorisation, and simple matching helpers.

expletives

Curated profanity lists with provenance, classification, and simple matching helpers, merged from 41 public wordlists across 15 languages.

pip install expletives

Usage

>>> import expletives
>>> expletives.is_expletive("fuck")
True
>>> expletives.is_expletive("asian")          # curated false-positive
False
>>> expletives.contains_expletive("hello world")
False
>>> expletives.find_expletives("this shit is fucked")
[Match(word='shit', start=5, end=9), Match(word='fucked', start=13, end=19)]
>>> expletives.censor("oh shit, fuck!")
'oh ****, ****!'
>>> expletives.censor("oh shit", mask="[BEEP]")
'oh [BEEP]'

Rich metadata for every entry:

>>> expletives.catalog["fuck"]["sources"][:3]
['2600-googleblacklist', 'bannedwordlist', 'biglou-bad-words']
>>> expletives.catalog["fuck"]["classification"]
'term'
>>> expletives.explain_okayish("asian")
'nationality / ethnicity / geographic descriptor'

Filter by source or language:

>>> expletives.load(sources=["seven-dirty-words"])
{'cocksucker', 'cunt', 'fuck', 'motherfucker', 'piss', 'shit', 'tits'}
>>> len(expletives.load(language="ja"))
180

Module attributes

name	type	description
badwords	set[str]	merged words, wildcards excluded
patterns	set[str]	wildcard entries (*fuck*, …)
okayish	list[str]	curated false-positive allow-list
catalog	dict[str, dict]	full entry metadata (see SOURCES.md)
sources	dict[str, dict]	per-source metadata
classifications	tuple[str, ...]	valid classification ids

Functions

function	returns
is_expletive(word, allow_okayish=True)	bool
contains_expletive(text)	bool
find_expletives(text)	list[Match]
censor(text, mask="*")	str
load(sources=None, language=None, ...)	set[str]
describe(word) / classify(word)	str | None
explain_okayish(word)	str | None

Sources & licensing

See SOURCES.md for every bundled wordlist, its origin, and sources we evaluated but didn't include.

Apache 2.0. The bundled source files retain their upstream licenses as noted in each file's header.