Struct NullLlmProvider 

pub struct NullLlmProvider;

No-op LLM provider for unit tests.

Use this when you need to build a ToolContext (which holds an Arc<dyn LlmProvider>) in test code, but the test itself does not exercise any LLM call path. Any call to chat_completion returns an empty success.

Trait Implementations

impl LlmProvider for NullLlmProvider

fn chat_completion<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, _messages: &'life1 [Message], _tools: &'life2 [ToolDefinition], ) -> Pin<Box<dyn Future<Output = Result<LlmResponse>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

fn is_copilot(&self) -> bool

Returns true when this provider is talking to GitHub Copilot. Used by the REPL to show auth status and guard /login.

fn set_stream_print(&self, _enabled: bool)

Enable / disable real-time streaming output to stdout. Called by Plan mode to suppress live streaming so the reply can be post-processed before display. No-op for providers that don’t stream.

fn set_copilot_oauth_token<'life0, 'async_trait>( &'life0 self, _token: String, ) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Update the long-lived OAuth token after /login completes. Only meaningful for the Copilot provider; all others do nothing.