logo

go version version version

Pyramid

Fast generic Heap in Golang. Create your Heap data structure with any data type. Set your compression function to sort your data with any field that you want. Push, Pop and Update your data in heap without any problems. Write your equality function to filter the elements that you want to update.

Pyramid is built with Golang internal libraries and does not include any external libraries.

How to use?

Get pyramid package.

go get github.com/amirhnajafiz/pyramid

Now you can use pyramid to build any type of Heap.

Example

Creating a max-heap of type Integer.

package main

import (
	"fmt"

	"github.com/amirhnajafiz/pyramid"
)

func main() {
	// initializing a heap with comparator function
	h := pyramid.NewHeap[int](func(a int, b int) bool {
		return a > b
	})

	// Using push method
	h.Push(2)
	h.Push(12)
	h.Push(4)
	h.Push(90)
	h.Push(20)

	// Using length method
	for h.Length() > 0 {
		// Using pop method
		fmt.Printf("%d ", h.Pop().(int))
	}
}

Creating a min-heap of type custom Data.

// Creating a custom Data struct type.
type Data struct {
	Priority int
	Data     string
}

func main() {
	// Creating a heap of Data type.
	h := pyramid.NewHeap[Data](func(a Data, b Data) bool {
		return a.Priority < b.Priority
	})

	// Push data into heap.
	for i := 0; i < 10; i++ {
		h.Push(Data{Priority: i, Data: fmt.Sprintf("data: %d", i+1)})
	}

	for h.Length() > 0 {
		fmt.Printf("%s\n", h.Pop().(Data).Data)
	}
}

Updating a reference

You can update any item in the list, also you can define a equal function to check the equality of your objects.

special := 675

h.Push(special)

for i := 2; i < 100; i++ {
    h.Push(i)
}

h.Update(special, 0, func(a int, b int) bool {
    return a == b
})

Load test

If we have 10000 items, and we want to update them 1 Million times, it would only take 12 seconds:

go run load-test/update/main.go -push 10000 -update 1000000

testing: 10000 numbers
start: Oct 16 14:42:34.752
done: Oct 16 14:42:52.064
final size: 10000
min: 2

GitHub

View Github