Efficient Data Structure for Graph Social Network using golang

main.go

package main
  
import (
	"encoding/json"
	"log"
	"net/http"
)

type Relationship struct {
	Rid  int
	Type string
	With int
}

type Person struct {
	Pid      int
	Name     string
	Relation map[string][]int
}

//-------------------------------------------------------------
func (p *Person) AddRelation(q *Person, name string) *Person {
	r := &Relationship{101, name, q.Pid}
  
	if p.Relation == nil {
		p.Relation = make(map[string][]int)
	}
	//Initializing map

	p.Relation[name] = append(p.Relation[name], r.Rid) //to relatinship id only
	return p
}

//-------------------------------------------------------------

func main() {
	//api request to be match with url SOMETHING LIKE CHAT INTERFACE
	http.HandleFunc("/", func(rw http.ResponseWriter, req *http.Request) {
		rw.Header().Add("Content-type", "application/json")

		p1 := &Person{Pid: 1}
		p1.Name = "P" //this id and other details may come from pages: our and other
		p2 := &Person{2, "S", nil}

		p1.AddRelation(p2, "WIFE")
		p1.AddRelation(p2, "FRIEND")
		p2.AddRelation(p1, "HUSBAND")

		b, err := json.Marshal(p1)
		if err != nil {
			log.Fatal(err)
		}
		rw.Write(b)
	})
	log.Println("Running on http://localhost:9890")
	log.Fatal(http.ListenAndServe(":9890", nil))
}

Efficient Design

• Person is mapped to ID of Relationship. Relationship is mapped to other Person. O(1) to get from P to R to P.
• Person.Relation is a map to array of int, so that Person.Relation[$relationName] will be fetched in O(1). And such such relations having same $relationName are Edges to given Person Node. ($relationName is variable here e.g. "FRIEND")

So this is 1-level query, now next level

• A Function will be present to map Person.Id and Relationship.Id to respective object data. (Note: Make these ID's as primary key for database so that fetching them will be in O(1) )
• This above function let's say idfy(Person.Id) will map to the person object. Now you can query within those person object as Person.Relation[$relationName2]. Complexity: O(E)

Level can grow like this with complexity of O(E).