org.llm4s.javaapi.reactor

A model's reply as a Reactor Flux<String> of its text (#1492), for Spring WebFlux and other Reactor pipelines.

Each method adapts the JDK Flow.Publisher of org.llm4s.javaapi.JLlmClient#streamAsPublisher with Reactor's JdkFlowAdapter, so the Flux behaves as that publisher does:

• it is cold: each subscription makes its own provider call, on a thread of its own;
• it emits the pieces of the reply's text in order, then completes;
• a failed call ends it with onError, carrying an LlmException whose getKind() says what failed;
• it honours backpressure: a subscriber that requests less than the model produces holds the provider's thread rather than losing text;
• cancelling the subscription (dispose(), take(n), a client that disconnects) interrupts the provider call.

A null argument is never thrown: the Flux fails on subscription with an LlmException of kind VALIDATION.

Flux<String> reply = Streams.toFlux(client, "Explain monads in one sentence");
reply.subscribe(System.out::print);