Struct FakeLlmProvider 

pub struct FakeLlmProvider { /* private fields */ }

Implementations

impl FakeLlmProvider

pub fn new(responses: Vec<Vec<LlmResponse>>) -> Self

pub fn with_single_response(chunks: Vec<LlmResponse>) -> Self

pub fn from_results(responses: Vec<Vec<Result<LlmResponse, LlmError>>>) -> Self

pub fn with_display_name(self, name: &str) -> Self

pub fn with_context_window(self, window: Option<u32>) -> Self

pub fn pause_turn_after( self, turn_index: usize, chunk_index: usize, notify: Arc<Notify>, ) -> Self

Pause the stream for turn_index after emitting the chunk at chunk_index until the returned Notify is notified. Enables deterministic mid-stream tests (e.g. send a user message after the provider has emitted some text).

pub fn captured_contexts(&self) -> Arc<Mutex<Vec<Context>>>

Returns a handle to the captured contexts that can be used to verify what contexts were passed to the LLM.

Trait Implementations

impl StreamingModelProvider for FakeLlmProvider

fn stream_response(&self, context: &Context) -> LlmResponseStream

fn display_name(&self) -> String

fn context_window(&self) -> Option<u32>

Context window size in tokens for the current model. Returns None for unknown models (e.g. Ollama, LlamaCpp).

fn model(&self) -> Option<LlmModel>

The LlmModel this provider is currently configured to use. Returns None for providers where the model is unknown at compile time (e.g. test fakes).