ShrtKt

Simple URL Shortener API written in Ktor 🚀

Features

• URL Shortening: Create short codes for long URLs
• Expiry Support: Set expiration dates for shortened URLs
• User Management: API key-based authentication
• Analytics: Track visit counts and top URLs
• Delete: URLs can be marked as deleted
• User Tiers: HOBBY and ENTERPRISE with tier-based access (bulk is enterprise-only)
• User Urls: GET /urls returns a list of the user's shortened URLs with longUrls.
• Bulk Shorten: POST /bulk/shorten accepts an array of requests and returns per-item results with a summary
• Custom Short Code: provide shortCode on create/update to set your own code
• Password Protection: optional password required at redirect
• ** Blacklist User**: Adds a user's API key to the blacklist.json file. Any API request using the same key will then be blocked from accessing any resources.

API Documentation

Test the API directly with Postman collection:

Or use Bruno API client with the collection in the bruno/ directory.

How to run the server

Run this command and server should be up, on port 8080

  ./gradlew clean run

Or with docker compose:

  docker compose up -d

This should return OK

  curl 0.0.0.0:8080/status

How to run tests

  ./gradlew clean test

Output

> Task :test

ApplicationTest > testRoot PASSED

ApplicationTest > testShortenAndRedirect PASSED

Sample API Keys

Use the following test API keys via the x-api-key header:

• Alice: sk_test_alice
• Bob: sk_test_bob

Expiry Support

URLs can have expiration dates.

• Expired URLs return 404 on redirect attempts.
• Expiry can be updated via PUT requests.
• Expired URLs are automatically filtered.

Format: YYYY-MM-DDTHH:MM:SS (ISO 8601)

API Performance

Test Configuration

• Endpoints: /shorten and /redirect APIs
• Environment: Docker Compose with resource constraints: 2 CPU cores and 4GB RAM

To run the benchmark:

  k6 run benchmark/benchmark.js -u 400 -i 400 # concurrent users/iterations count

Cache performance benchmarks are available in the benchmark/ directory. Run benchmark/compare-cache-performance.sh to compare performance with and without cache.

Latency:

Database	Concurrent users	p(50)	p(90)	p(95)	p(99)	failure %
In-memory	10	6ms	9.01ms	9.93ms	10.51ms	0%
In-memory	50	15.78ms	28.39ms	29.61ms	36.99ms	0%
In-memory	100	18.84ms	39.96ms	40.37ms	40.89ms	0%
In-memory	200	20ms	46.45ms	47.6ms	53.47ms	0%
In-memory	400	32.79ms	65.05ms	67.37ms	76.23ms	0%
In-memory	800	44.84ms	79.65ms	84.73ms	94.47ms	0%
In-memory	1600	59.73ms	94.25ms	120.6ms	128.97ms	0%
In-memory	3200	299.1ms	880.01ms	1.03s	1.05s	0%
In-memory	6400	450.01ms	1.09s	1.14s	2.92s	0%
In-memory	12800	798.58ms	2.42s	3.13s	4.33s	0%
In-memory	13000	602.45ms	2.43s	5.11s	5.44s	0.05%