Metrics

Collects span timing data (p50/p95/p99) for performance monitoring.

{ metrics: true } (simple)

Add metrics: true to any config object. A per-logger MetricsCollector is created automatically and exposed as log.metrics.

import { createLogger } from "loggily"

const log = createLogger("myapp", [{ level: "debug", metrics: true }, console])

{
  using span = log.span("query")
  // ...
}

log.metrics.stats("myapp:query") // SpanStats | undefined
log.metrics.summary() // formatted string
log.metrics.all() // Map<string, SpanStats>

Child loggers inherit the same collector:

const db = log.child("db")
{
  using span = db.span("query")
  // ...
}
log.metrics.stats("myapp:db:query") // same collector

withMetrics(collector) (advanced)

Import from loggily/metrics. Use when you need a shared collector across multiple loggers or custom configuration.

import { withMetrics, createMetricsCollector } from "loggily/metrics"
import { createLogger } from "loggily"

const collector = createMetricsCollector()
const log = withMetrics(collector)(createLogger("myapp"))

// log.metrics === collector

{
  using span = log.span("query")
  // ...
}

collector.stats("myapp:query") // { count, min, max, mean, p50, p95, p99, total }
collector.summary() // formatted string
collector.reset() // clear data

API

Export	Description
{ metrics: true }	Config key: auto-creates collector on log.metrics
createMetricsCollector(maxEntries?)	Create a standalone collector (default 1000 entries per span)
withMetrics(collector)	Wrap a logger to record spans to a collector

SpanStats

interface SpanStats {
  count: number
  min: number
  max: number
  mean: number
  p50: number
  p95: number
  p99: number
  total: number
}