Canoe Test

Gettings Started

It's need to install in your computer: Docker: 24.0.0+

Execute this commands to start the application:

// Create the .env file
cp .env.example .env
// Start all containers docker
make up
// Enter in container
make bash
// Generate the key
php artisan key:generate
// It's everything!
// You can access the api in: http://localhost:8040/

Framework

Laravel

Database

MySQL 8.0
Redis 6.2

ER Diagram

Migrations

// Create migrations (First time)
php artisan migrate --seed
// Drop/Create migrations with seeds
php artisan migrate:fresh --seed
// Install database without datas
php artisan migrate:fresh

API Docs

There is a postman collection in the root of the project called docs/canoe.postman_collection.json

Get all funds

curl --location --request GET 'http://localhost:8040/api/funds'

Get all funds with filters

curl --location --request GET 'http://localhost:8040/api/funds?search=name%3ABergstrom%3Bstart_year%3A1987%3Bmanager.name%3AParisian'

Get fund by id

curl --location --request GET 'http://localhost:8040/api/funds/1'

Create fund

curl --location --request POST 'http://localhost:8040/api/funds' \
--header 'Content-Type: application/json' \
--data '{
    "name": "Fund name",
    "start_year": "2023",
    "manager_id": 1
}'

The listener will write a log message in the file storage/logs/laravel-d-m-Y.log when receive a event FundDuplicateWarning from the event dispatcher.

Update fund

curl --location --request PUT 'http://localhost:8040/api/funds/1' \
--header 'Content-Type: application/json' \
--data '{
    "name": "Fund name",
    "start_year": "2023",
    "manager_id": 1
}'

Delete fund

curl --location --request DELETE 'http://localhost:8040/api/funds/1'

Tests

// Run all tests
make bash
php artisan test

Scalability considerations:

How will your application work as the data set grows increasingly larger?

Will work fine, but we can improve the performance with the following actions:

Optimize the database queries with proper indexing
Use a cache system like Redis to store the data and avoid database queries
Use a database cluster to distribute the database queries between multiple instances

How will your application work as the # of concurrent users grows increasingly larger?