Struct CompressionAggregateCache 

pub struct CompressionAggregateCache { /* private fields */ }

Process-local compression totals backed by the durable event table.

Status reads validate entries with the table’s maximum row id, an indexed lookup that detects writes from other AFT processes. Full aggregate scans run only for a cold or stale key. Successful local inserts advance warm entries directly while the caller still owns the database connection mutex.

Implementations

impl CompressionAggregateCache

pub fn aggregates_for_session( &self, conn: &Connection, harness: &str, project_key: &str, session_id: &str, ) -> Result<(CompressionAggregate, CompressionAggregate)>

pub fn record_successful_insert( &self, conn: &Connection, row: &CompressionEventRow<'_>, inserted_row_id: i64, )

Apply a row that was inserted successfully on conn.

A warm entry is advanced only when its watermark matches the row that immediately preceded inserted_row_id. If another process wrote first, the entry remains stale and the next status read rebuilds it from SQL.

pub fn clear(&self)

Trait Implementations

impl Debug for CompressionAggregateCache

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

Formats the value using the given formatter.

impl Default for CompressionAggregateCache

fn default() -> CompressionAggregateCache

Returns the “default value” for a type.