Struct TokenTracker 

pub struct TokenTracker {
    pub provider_id: Option<String>,
    pub model_id: Option<String>,
    pub context_limit: Option<u64>,
    pub latest_input_tokens: Option<u64>,
    pub compaction_needed: bool,
    /* private fields */
}

Tracks token usage during a session run to detect context limit threshold.

Fields

provider_id: Option<String>

Provider ID from message events

model_id: Option<String>

Model ID from message events

context_limit: Option<u64>

Context limit for the current model (from cached limits)

latest_input_tokens: Option<u64>

Latest observed input token count

compaction_needed: bool

Flag indicating compaction/summarization is needed

Implementations

impl TokenTracker

pub fn with_threshold(threshold: f64) -> Self

Create a new token tracker with a custom compaction threshold.

The threshold should be between 0.0 and 1.0 (e.g., 0.80 for 80%).

pub fn observe_event<F>(&mut self, ev: &Event, context_limit_lookup: F)

where F: Fn(&str, &str) -> Option<u64>,

Observe an SSE event and update token tracking.

The context_limit_lookup function is called to look up the context limit for a given (provider_id, model_id) pair from the cached limits.

pub fn observe_tokens(&mut self, tokens: &TokenUsage)

Observe token usage and update threshold flag.

Trait Implementations

impl Clone for TokenTracker

fn clone(&self) -> TokenTracker

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for TokenTracker

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for TokenTracker

fn default() -> Self

Returns the “default value” for a type.