KSUID (K-Sortable Unique IDentifier)

KSUID:

KSUID is for K-Sortable Unique IDentifier. It is a kind of globally unique identifier similar to a RFC 4122 UUID, built from the ground-up to be "naturally" sorted by generation timestamp without any special type-aware logic.

It was originally open-sourced as a Go Library by Segment.

A sorteable unique identifier is useful in scenarios where you need your primary keys to be sorteable. For example, if you need to generate keys for your items in DynamoDB. In fact, I learned about KSUID while reading DynamoDB Book

There is a Java implementation of Segment's KSUID library. You can generate a KSUID String with: new Ksuid().generate().

For example, if you have a Contact POJO:

package example;

import com.amirkhawaja.Ksuid;

import java.io.IOException;

public class Contact implements Comparable<Contact> {

    private final String id;

    private final String name;

    private final String title;

    public Contact(String name, String title) throws IOException {
        this.id = new Ksuid().generate();
        this.name = name;
        this.title = title;
    }

    @Override
    public int compareTo(Contact o) {
        return o.getId().compareTo(getId());
    }

    public String getName() {
        return name;
    }

    public String getTitle() {
        return title;
    }

    public String getId() {
        return id;
    }

    @Override
    public String toString() {
        return "Contact{" +
                "id='" +  id + " " + new Ksuid().parse(id) + '\'' +
                ", name='" + name + '\'' +
                ", title='" + title + '\'' +
                '}';
    }

    @Override
    public boolean equals(Object o) {
        if (this == o) return true;
        if (o == null || getClass() != o.getClass()) return false;

        Contact contact = (Contact) o;

        if (id != null ? !id.equals(contact.id) : contact.id != null) return false;
        if (name != null ? !name.equals(contact.name) : contact.name != null) return false;
        return title != null ? title.equals(contact.title) : contact.title == null;
    }

    @Override
    public int hashCode() {
        int result = id != null ? id.hashCode() : 0;
        result = 31 * result + (name != null ? name.hashCode() : 0);
        result = 31 * result + (title != null ? title.hashCode() : 0);
        return result;
    }
}

You can sort it via the primary key:

Contact tim = new Contact("Tim Cook", "CEO");
Contact katherine = new Contact("Katherine Adams", "Senior Vice President and General Counsel");
Contact eddy = new Contact("Eddy Cue", "Senior Vice President Internet Software and Services");
Contact craig = new Contact("Craig Federighi", "Senior Vice President Software Engineering");
List<Contact> contacts = new ArrayList<>();
contacts.add(craig);
contacts.add(eddy);
contacts.add(tim);
contacts.add(katherine);
            
Collections.sort(contacts);

assert contacts.get(0).equals(tim);
System.out.println(tim.toString());

The primary key contains the timestamp. The previous code prints:

Contact{id='DRIvC0YHDXBuGawemfYsenzBN84 Time: 2021-04-24T20:18:19Z[UTC]
Timestamp: 1619295499
Payload: [70, 7, 13, 112, 110, 25, -84, 30, -103, -10, 44, 122, 124, -63, 55, -50]', name='Tim Cook', title='CEO'}

Tags: #java #ksuid
Apr 2021, 24.