[vyhuholl]

Description

Part of #3202.

ListBatchPublisher subclassed ListPublisher but replaced publish(message, list, reply_to, ...) with publish(*messages, list=..., ...). mypy accepts that override, since every positional argument still fits *messages, but the meaning changes. Code written against ListPublisher type checks a call that silently does something else on a batch publisher:

async def send(pub: ListPublisher) -> None:
    await pub.publish("hello", "other")  # "hello" to the list "other"


pub = broker.publisher(list=ListSub("default", batch=True))
await send(pub)
# on a ListBatchPublisher: ["hello", "other"] goes to "default" as one batch

It also applies without an explicit annotation: publisher(list=ListSub(q, batch=flag)) returns ListPublisher | ListBatchPublisher, and the same call type checks on that union. The snippet above behaves this way under TestRedisBroker.

Among pyrefly's 104 bad-override* findings on faststream/, this one (redis/publisher/usecase.py:263) is the only override a user can hit. The rest are summarized in #3202.

Fix

What the two publishers share (the constructor, list, subscriber_property, request) moves to _ListPublisherMixin, and ListPublisher and ListBatchPublisher both derive from it. The list subscribers are already split this way through _ListHandlerMixin, and Kafka's DefaultPublisher / BatchPublisher are siblings too. No method changes.

What changes for users: isinstance(batch_publisher, ListPublisher) is now False, and passing a ListBatchPublisher where ListPublisher is annotated is a type error. Nothing in faststream/, tests/, docs/docs_src/ or examples/ relies on either.

Tests

tests/mypy/redis.py passes a batch publisher to a function annotated with ListPublisher, under # type: ignore[arg-type]. On the old hierarchy mypy fails it with unused-ignore. Runtime behaviour isn't changed, so there's no runtime test.

Type of change

Please delete options that are not relevant.

• Bug fix (a non-breaking change that resolves an issue)

Checklist

• My code adheres to the style guidelines of this project (just lint shows no errors)
• I have conducted a self-review of my own code
• My changes do not generate any new warnings
• I have added tests to validate the effectiveness of my fix or the functionality of my new feature
• Both new and existing unit tests pass successfully on my local environment by running just test-coverage
• I have ensured that static analysis tests are passing by running just static-analysis