:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.
Python ab-testingai-analyticsanalyticscdpdata-warehouseexperimentsfeature-flagsjavascriptproduct-analyticspythonreactsession-replaysurveystypescriptweb-analytics
import { PersonsStoreForBatch } from '~/ingestion/common/persons/persons-store-for-batch'
import { PipelineResult, ok } from '~/ingestion/framework/results'
import { PipelineEvent, Team } from '~/types'
type PrefetchPersonsStepInput = { event: PipelineEvent; team: Team; personsStoreForBatch: PersonsStoreForBatch }
export function prefetchPersonsStep<T extends PrefetchPersonsStepInput>(enabled: boolean) {
return function prefetchPersonsStep(events: T[]): Promise<PipelineResult<T>[]> {
if (enabled && events.length > 0) {
// Events in a chunk may come from different Kafka batches (due to the feed primitive).
// Group by batch store so each store only receives entries it owns. Fire without
// awaiting — fetchForChecking/fetchForUpdate will wait on the pending promises if
// they need data that's still being fetched.
const entriesByStore = new Map<
PersonsStoreForBatch,
{ teamId: number; distinctId: string; batchId: number }[]
>()
for (const event of events) {
let entries = entriesByStore.get(event.personsStoreForBatch)
if (!entries) {
entries = []
entriesByStore.set(event.personsStoreForBatch, entries)
}
entries.push({
teamId: event.team.id,
distinctId: event.event.distinct_id,
batchId: event.personsStoreForBatch.batchId,
})
}
for (const [store, entries] of entriesByStore) {
void store.prefetchPersons(entries)
}
}
return Promise.resolve(events.map((event) => ok(event)))
}
}