hashmap

An efficient hashmap implementation in Go.

Features

• Support for Generics.
• Map and Set types for unordered key-value maps and sets,
• xxh3 algorithm
• Open addressing with Robin hood hashing
• Automatically shinks memory on deletes (no memory leaks).
• Pretty darn good performance. 🚀 (benchmarks).

For ordered key-value data, check out the tidwall/btree package.

Getting Started

Installing

To start using hashmap, install Go and run go get:

go get github.com/tidwall/hashmap

This will retrieve the library.

Usage

The Map type works similar to a standard Go map, and includes the methods: Set, Get, Delete, Len, Scan, Keys, Values, and Copy.

var m hashmap.Map[string, string]
m.Set("Hello", "Dolly!")
val, _ := m.Get("Hello")
fmt.Printf("%v\n", val)
val, _ = m.Delete("Hello")
fmt.Printf("%v\n", val)
val, _ = m.Get("Hello")
fmt.Printf("%v\n", val)

// Output:
// Dolly!
// Dolly!
//

The Set type is like Map but only for keys. It includes the methods: Insert, Contains, Delete, Len, Scan and Keys.

var m hashmap.Set[string]
m.Insert("Andy")
m.Insert("Kate")
m.Insert("Janet")

fmt.Printf("%v\n", m.Contains("Kate"))
fmt.Printf("%v\n", m.Contains("Bob"))
fmt.Printf("%v\n", m.Contains("Andy"))

// Output:
// true
// false
// true

Performance

See BENCH.md for more info.

Contact

Josh Baker @tidwall

License

Source code is available under the MIT License.