Parameters table

Summary of all parameters.

✅ means that parameter acts when RandomGenerator is created: it either configures the generator, or acts as load-time validation. If such validation fails, ConfigurationError is raised.

🚀 means that parameter acts at runtime every time generate() or generate_slug() is called. It’s to silently discard unwanted combinations.

Parameter	Scope	When	Type	Default value	Description
allow_whitespace	Words list, Phrases list, *.txt, Constant list; "all" list as a fallback [1]	✅	bool	False	Allow words or words within a phrase to contain whitespace.
generator	Any top-level list	✅	bool	False [2]	List is a generator endpoint and can be used in generate().
ensure_unique	RandomGenerator instance via "all" list	🚀	bool	True	Don’t generate combinations with repeating words.
ensure_unique_prefix	RandomGenerator instance via "all" list	🚀	int | None	None	Don’t generate combinations where first N symbols of any two word match, e.g. "great-fox-of-greatness".
max_length	Words list, Phrases list, *.txt	✅	int | None	None	Maximum number of characters in a word or phrase (not counting separators).
max_slug_length	RandomGenerator instance via "all" list	🚀	int | None	None	Don’t generate slugs with total length (including separators) exceeding N.
number_of_words	Phrases list, *.txt	✅	int | None	None	Exact number of words in every phrase in this list.
separator	Phrases list, *.txt; "all" list as a fallback [1]	✅	str	r're:\s+'	Separator used to split phrases defined as plain strings (not as lists/tuples). Use 're:' prefix for regular expression.
strip_whitespace	Words list, Phrases list, *.txt; "all" list as a fallback [1]	✅	bool	True	Strip leading and trailing whitespace from phrases and words before further processing.
word_regex	Words list, Phrases list, *.txt, Constant list; "all" list as a fallback [1]	✅	str	r'\w+'	Regex to validate all words. Default: any Unicode letters, numbers, underscore. [3]

Footnotes

[1] (1,2,3,4)

strip_whitespace, allow_whitespace, separator and word_regex can be defined at "all" level (affects all Word/Phrase/Constant lists, lower precedence) or at individual list level (higher precedence).

[2]

Special cases where default is generator=True: main "all" list and lists with integer names (such as "4").

[3]

If allow_whitespace=True, then word_regex is applied only to non-whitespace parts of the word. You can include whitespace in the regex in such configuration (like [a-z ]+), but you don’t have to ([a-z]+ is fine).