pg_slug_gen

Generate cryptographically secure timestamp-based slugs

Overview

PackageVersionCategoryLicenseLanguage
pg_slug_gen1.0.0FUNCMITC
IDExtensionBinLibLoadCreateTrustRelocSchema
4550pg_slug_genNoYesNoYesNoYes-
Relatedpg_hashids sequential_uuids uuid-ossp pg_uuidv7

Version

TypeRepoVersionPG VerPackageDeps
EXTPIGSTY1.0.01817161514pg_slug_gen-
RPMPIGSTY1.0.01817161514pg_slug_gen_$v-
DEBPIGSTY1.0.01817161514postgresql-$v-pg-slug-gen-
OS / PGPG18PG17PG16PG15PG14
el8.x86_64PIGSTY MISS
el8.aarch64PIGSTY MISS
el9.x86_64PIGSTY MISS
el9.aarch64PIGSTY MISS
el10.x86_64PIGSTY MISS
el10.aarch64PIGSTY MISS
d12.x86_64PIGSTY MISS
d12.aarch64
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY MISS
d13.x86_64
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY MISS
d13.aarch64
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY MISS
u22.x86_64
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY MISS
u22.aarch64
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY MISS
u24.x86_64
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY MISS
u24.aarch64
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY 1.0.0
PIGSTY MISS

Build

You can build the RPM / DEB packages for pg_slug_gen using pig build:

pig build pkg pg_slug_gen         # build RPM / DEB packages

Install

You can install pg_slug_gen directly. First, make sure the PGDG and PIGSTY repositories are added and enabled:

pig repo add pgsql -u          # Add repo and update cache

Install the extension using pig or apt/yum/dnf:

pig install pg_slug_gen;          # Install for current active PG version
pig ext install -y pg_slug_gen -v 18  # PG 18
pig ext install -y pg_slug_gen -v 17  # PG 17
pig ext install -y pg_slug_gen -v 16  # PG 16
pig ext install -y pg_slug_gen -v 15  # PG 15
dnf install -y pg_slug_gen_18       # PG 18
dnf install -y pg_slug_gen_17       # PG 17
dnf install -y pg_slug_gen_16       # PG 16
dnf install -y pg_slug_gen_15       # PG 15
apt install -y postgresql-18-pg-slug-gen   # PG 18
apt install -y postgresql-17-pg-slug-gen   # PG 17
apt install -y postgresql-16-pg-slug-gen   # PG 16
apt install -y postgresql-15-pg-slug-gen   # PG 15

Create Extension:

CREATE EXTENSION pg_slug_gen;

Usage

Syntax:

CREATE EXTENSION pg_slug_gen;
SELECT gen_random_slug();
SELECT gen_random_slug(13);

Source: PGXN release README

pg_slug_gen generates timestamp-based slugs using cryptographic randomness. The PGXN release README describes it as a PostgreSQL extension that maps timestamp digits into letter buckets and inserts a hyphen in the middle, producing URL-friendly slugs.

Function

The documented SQL function is:

gen_random_slug(slug_length int DEFAULT 16) RETURNS text

The README shows these interfaces:

gen_random_slug()      -- default: 16 (microseconds)
gen_random_slug(10)    -- seconds
gen_random_slug(13)    -- milliseconds
gen_random_slug(16)    -- microseconds
gen_random_slug(19)    -- nanoseconds

Precision and Length

The release README maps precision to timestamp digits and maximum collision-free throughput:

  • 10 digits for seconds, up to 1 insert per second
  • 13 digits for milliseconds, up to 1,000 inserts per second
  • 16 digits for microseconds, up to 1,000,000 inserts per second
  • 19 digits for nanoseconds, up to 1 billion inserts per second

The slug includes a midpoint hyphen:

  • seconds: 5-5 pattern, 11 characters total
  • milliseconds: 6-7 pattern, 14 characters
  • microseconds: 8-8 pattern, 17 characters
  • nanoseconds: 9-10 pattern, 20 characters

Examples

SELECT gen_random_slug();
SELECT gen_random_slug(10);
SELECT gen_random_slug(13);
SELECT gen_random_slug(16);
SELECT gen_random_slug(19);

As a table default:

CREATE TABLE products (
    id serial PRIMARY KEY,
    name text NOT NULL,
    slug text DEFAULT gen_random_slug() UNIQUE
);

How It Works

The release README describes the algorithm as:

  1. take the current timestamp at the chosen precision
  2. map each digit to a QWERTY-based character bucket
  3. choose one random character from that bucket using pg_strong_random()
  4. insert a hyphen at the midpoint

The README also notes that same-timestamp collisions remain possible, but with microsecond precision the probability is stated as roughly 1 in 10 million.


Last Modified 2026-04-14: update extension catalog (29617e5)