Struct TokenStatsAggregator 

pub struct TokenStatsAggregator { /* private fields */ }

In-memory aggregator for batched token daily stats updates.

During batch ingestion, accumulate token deltas per (day_id, agent, source, model_family) key. After processing, call expand() to generate the 5 permutation keys, then flush via update_token_daily_stats_batched_in_tx.

Implementations

impl TokenStatsAggregator

pub fn new() -> Self

pub fn record( &mut self, agent_slug: &str, source_id: &str, day_id: i64, model_family: &str, role: &str, usage: &ExtractedTokenUsage, content_chars: i64, estimated_cost_usd: f64, )

Record a single message’s token contribution.

pub fn record_session( &mut self, agent_slug: &str, source_id: &str, day_id: i64, model_family: &str, )

Record a session count bump for a given day/agent/source/model.

pub fn expand(&self) -> Vec<(i64, String, String, String, TokenStatsDelta)>

Expand raw deltas into 5 permutation keys for the 4-dimensional composite PK:

• (agent, source, model) — specific all three
• (“all”, source, model) — all agents
• (agent, “all”, model) — all sources
• (agent, source, “all”) — all models
• (“all”, “all”, “all”) — global total

pub fn is_empty(&self) -> bool

pub fn raw_entry_count(&self) -> usize

Trait Implementations

impl Debug for TokenStatsAggregator

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

Formats the value using the given formatter.

impl Default for TokenStatsAggregator

fn default() -> TokenStatsAggregator

Returns the “default value” for a type.