Struct ServerState 

pub struct ServerState { /* private fields */ }

Central server state tracking provider configuration and cached runners

Runners are created lazily on first access and cached for reuse. The active provider and model determine how prompt dispatch behaves.

Implementations

impl ServerState

pub fn new(default_provider: CliRunnerType) -> ServerState

Create server state with the given default provider

pub async fn active_provider(&self) -> CliRunnerType

Get the currently active provider type

pub async fn set_active_provider(&self, provider: CliRunnerType)

Switch the active provider (resets the active model)

pub async fn active_model(&self) -> Option<String>

Get the currently selected model (None means use provider default)

pub async fn set_active_model(&self, model: Option<String>)

Set the model to use for subsequent requests

pub async fn multiplex_providers(&self) -> Vec<CliRunnerType>

Get the list of providers configured for multiplex dispatch

pub async fn set_multiplex_providers(&self, providers: Vec<CliRunnerType>)

Set the providers used when multiplexing prompts

pub async fn get_runner( &self, provider: CliRunnerType, ) -> Result<Arc<dyn LlmProvider>, RunnerError>

Get or lazily create a runner for the given provider type

Created runners are cached for future calls. The runner cache uses interior mutability so callers only need &self.