Swapi-caching

all credits to https://github.com/matt-jb/

Prereqs

• npm
• docker
• valid env file (for testing purposes, you may just rename the .env.example file to .env)

Run project

App consists of Nest App that connects to PostgreSQL database (v14).

Run app:

$ docker-compose up -d

Validate endpoint health:

$ curl http://localhost:3000/health

> StatusCode        : 200
> StatusDescription : OK
> Content           : {"status":"OK"}
> RawContent        : HTTP/1.1 200 OK

Stop application:

$ docker-compose down

Endpoints

POST /swapi-query/page

For retreiving full pages content.

The content will cache for 24 hours, so any request for the same data before that time will result in serving content cached in the PostgreSQL database.

Request body:

resource!: enum ("films" | "planets" | "species" | "starships" | "vehicles")
page?: number

Response example:

{
    "count": 60,
    "next": "https://swapi.dev/api/planets/?page=2",
    "previous": null,
    "results": [
        {
            "name": "Tatooine",
...
}

POST /swapi-query/item

For retreiving data about single entities (films, planets, etc.).

The content will cache for 24 hours, so any request for the same data before that time will result in serving content cached in the PostgreSQL database.

Request body:

resource!: enum ("films" | "planets" | "species" | "starships" | "vehicles")
id!: number

Response example:

{
    "name": "Alderaan",
    "rotation_period": "24",
    "orbital_period": "364",
    "diameter": "12500",
...
}

GET /stats

For retreiving statistics about:

1. Characters most frequently mentioned in the films' opening crawls.
2. Words counts in opening crawls.

The content will cache for 24 hours, so any request for the same data before that time will result in serving content cached in the PostgreSQL database.

Response example:

{
    "charactersStatistics": {
        "description": "Most frequent characters in opening crawls",
        "data": [
            "Luke Skywalker",
            "Dooku"
        ]
    },
    "wordsStatistics": {
        "description": "Words counts in opening crawls",
        "data": {
            "it": 3,
            "is": 7,
            "a": 9,
            "period": 1,
            "of": 21,
...
}

PUT /stats/update

Manually triggering a refresh of the data for the GET /stats route.

If data isn't cached (or the cache is not fresh), the GET /stats route sends multiple requests to the /people SWAPI endpoint (and a single /films endpoint request). Since SWAPI is quite slow to respond, this results in returning fresh data only after 10+ seconds. This sort of route should receive, e.g. daily, a CRON job request for refreshing the cache. This would make this long operation always having fresh data and making response times instantaneous.

IMPORTANT: No body is needed for this endpoint, but you need to provide a valid X-Stats-Auth header in order to be validated (default value in the .env.example file is: foo-bar).