prime_list

prime_list(n, base_primes=None, piter=None): 

LLM Docstring

Return the first n primes, using a shared incremental cache by default.

Three cases, chosen by what’s supplied:

• piter=None, base_primes=None (the default call): both are pulled from the module-level default_prime_iter/default_base_prime_list, a cache that’s intentionally persistent across calls.
• piter=None, base_primes=<a list>: a fresh iterator is built from that list with prime_iter(base_primes), instead of pulling from the unrelated global cache. Previously, supplying a custom base_primes without also supplying a matching piter silently ignored the supplied list’s content – the shared global generator’s cache won regardless, so the “custom starting list” argument didn’t do what it looked like it did.
• piter=<an iterator>: used as given; base_primes defaults to a new empty list if not also supplied (so it isn’t tied to the global cache unless the caller asks for that explicitly).

In every case base_primes is still extended in place and returned; supplying your own list lets you keep your own independent cache instead of sharing the module-level one.

• n: int

  number of primes requested

• base_primes: list[int] | None

  mutable cache populated in place; None selects the shared default cache (or, if piter is supplied, a fresh empty list)

• piter: collections.abc.Iterator[list[int]] | None

  cumulative prime-list iterator used to extend the cache; None selects the shared default iterator, unless base_primes was supplied, in which case a fresh iterator seeded from it is used

• :returns: list[int]

  the first n cached primes