: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 { Counter, Histogram, Meter, MetricOptions } from '@opentelemetry/api'
import { counterWithExemplars, histogramWithExemplars } from './exemplars'
/**
* Shared helpers for the OTLP twin modules (logs, session replay ingestion, rasterizer),
* so every instrument gets the exemplar wrap and every record path swallows telemetry
* errors the same way.
*/
// The wrappers buffer an exemplar (active span context) per record, which the
// OTLP JSON exporter attaches at export — the upstream SDK path can't do this.
export const createCounterWithExemplars = (meter: Meter, name: string, options?: MetricOptions): Counter =>
counterWithExemplars(name, meter.createCounter(name, options))
export const createHistogramWithExemplars = (meter: Meter, name: string, options?: MetricOptions): Histogram =>
histogramWithExemplars(name, meter.createHistogram(name, options))
/**
* Twin recording runs in hot paths, finally blocks, and error handlers. A throw here
* would mask the real processing error, so recording failures are swallowed.
*/
export function swallowing<Args extends unknown[]>(record: (...args: Args) => void): (...args: Args) => void {
return (...args: Args): void => {
try {
record(...args)
} catch {
// never let telemetry break the caller
}
}
}