evertqin/zap

Fast, structured, leveled logging in Go.

⚡ zap

Blazing fast, structured, leveled logging in Go.

Installation

go get -u go.uber.org/zap

Quick Start

In contexts where performance is nice, but not critical, use the SugaredLogger. It's 4-10x faster than than other structured logging libraries and includes both structured and printf-style APIs.

logger, _ := zap.NewProduction()
sugar := logger.Sugar()
sugar.Infow("Failed to fetch URL.",
  // Structured context as loosely-typed key-value pairs.
  "url", url,
  "attempt", retryNum,
  "backoff", time.Second,
)
sugar.Infof("Failed to fetch URL: %s", url)

When performance and type safety are critical, use the Logger. It's even faster than the SugaredLogger and allocates far less, but it only supports structured logging.

logger, _ := zap.NewProduction()
logger.Info("Failed to fetch URL.",
  // Structured context as strongly-typed Field values.
  zap.String("url", url),
  zap.Int("attempt", tryNum),
  zap.Duration("backoff", time.Second),
)

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 and fmt.Fprintf to log tons of interface{}s makes your application slow.

Zap takes a different approach. It includes a reflection-free, zero-allocation JSON encoder, and the base Logger strives to avoid serialization overhead and allocations wherever possible. By building the high-level SugaredLogger on that foundation, zap lets users choose when they need to count every allocation and when they'd prefer a more familiar, loosely-typed API.

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.¹

Log a message and 10 fields:

Library	Time	Bytes Allocated	Objects Allocated
⚡ zap	1466 ns/op	705 B/op	2 allocs/op
⚡ zap (sugared)	2893 ns/op	1931 B/op	21 allocs/op
go-kit	8183 ns/op	3119 B/op	65 allocs/op
lion	12259 ns/op	5999 B/op	62 allocs/op
logrus	12862 ns/op	5783 B/op	77 allocs/op
apex/log	20317 ns/op	4024 B/op	64 allocs/op
log15	31855 ns/op	5536 B/op	91 allocs/op

Log a message with a logger that already has 10 fields of context:

Library	Time	Bytes Allocated	Objects Allocated
⚡ zap	536 ns/op	0 B/op	0 allocs/op
⚡ zap (sugared)	734 ns/op	80 B/op	2 allocs/op
lion	6784 ns/op	3978 B/op	36 allocs/op
go-kit	8316 ns/op	2950 B/op	50 allocs/op
logrus	10160 ns/op	3967 B/op	61 allocs/op
apex/log	17095 ns/op	2801 B/op	49 allocs/op
log15	19112 ns/op	2545 B/op	42 allocs/op

Log a static string, without any context or printf-style templating:

Library	Time	Bytes Allocated	Objects Allocated
⚡ zap	521 ns/op	0 B/op	0 allocs/op
standard library	580 ns/op	80 B/op	2 allocs/op
⚡ zap (sugared)	885 ns/op	80 B/op	2 allocs/op
go-kit	1384 ns/op	656 B/op	13 allocs/op
lion	2009 ns/op	1224 B/op	10 allocs/op
logrus	2925 ns/op	1409 B/op	25 allocs/op
apex/log	3723 ns/op	584 B/op	11 allocs/op
log15	6349 ns/op	1496 B/op	24 allocs/op

Development Status: Stable

All APIs are finalized, and no breaking changes will be made in the 1.x series of releases. Users of semver-aware dependency management systems should pin zap to ^1.

---

Released under the MIT License.

¹ In particular, keep in mind that we may be benchmarking against slightly older versions of other libraries. Versions are pinned in zap's glide.lock file. ↩