⚡ zap

Fast, structured, leveled logging in Go.

Structure

Zap takes an opinionated stance on logging and doesn't provide any printf-style helpers. Rather than logger.Printf("Error %v writing logs to %v, lost %v messages.", err, f, m), zap encourages the more structured

Logger.
  WithError(err).
  With("msgCount", m).
  With("fileName", f).
  Info("Error writing logs.")

This a bit more verbose, but it enables powerful ad-hoc analysis, flexible dashboarding, and accurate message bucketing. In short, it helps you get the most out of tools like ELK, Splunk, and Sentry. All log messages are JSON-serialized.

Performance

For applications that log in the hot path, reflection-based serialization and string formatting are prohibitively expensive — they're CPU-intensive and make many small allocations. Put differently, using encoding/json to log tons of interface{}s makes your application slow.

Zap's API offers a variety of type-safe ways to annotate a logger's context without incurring tons of overhead. It also offers a suite of conditional annotations, so collecting rich debugging context doesn't impact normal operations.

As measured by its own benchmarking suite, not only is zap more performant than comparable structured logging libraries — it's also faster than the standard library. Like all benchmarks, take these with a grain of salt.

Add 5 fields to the logging context, one at a time:

Library	Time	Bytes Allocated	Objects Allocated
zap	3340 ns/op	5713 B/op	16 allocs/op
logrus	66776 ns/op	52646 B/op	254 allocs/op

Add 5 fields to the logging context as a single map:

Library	Time	Bytes Allocated	Objects Allocated
zap	1615 ns/op	1504 B/op	6 allocs/op
logrus	36592 ns/op	21409 B/op	209 allocs/op

Log static strings, without any context or printf-style formatting:

Library	Time	Bytes Allocated	Objects Allocated
zap	328 ns/op	32 B/op	1 allocs/op
standard library	840 ns/op	592 B/op	2 allocs/op

Development Status: Alpha

Breaking changes are certain.

---

Released under the [MIT License](LICENSE.txt).