FastAPI + Postgres + PgAmdin + Postgis app

Run the app with Docker

• Clone this repo

git clone https://github.com/jvelazquez-reyes/edt-test.git

• Create an .env file to store enviroment variables for Postgres-Postgis database and PgAmdin panel

DB_USER=postgres
DB_PASSWORD=password
DB_NAME=restaurant_db

[email protected]
PGADMIN_PASSWORD=admin

• Run the docker-compose.yml file:

docker-compose up

• Go to port 5050 to use the PgAmdin panel:

http://127.0.0.1:5050

• Login to PgAming using the credentials in the .env file:

[email protected]
PGADMIN_PASSWORD=admin

• Locate the navbar in the PgAmdin panel and go to Object > Register > Server. It will open a window to create a new server. Fill in the fields Host name/Address: db, Port: 5432, Username: postgres, and Password: password:

• Restart the app service. You can do this using the Docker Desktop. Locate the app service, hit the three dots on the right and restart the service as shown in the image below:

• Visit port 8000 to use the FastAPI app:

http://0.0.0.0:8000

• When visiting the previous endpoint you should see get the following response:

{"message":"Hello EDT Test"}

Endpoints

• Visit the built-in API docs from FastAPI:

http://0.0.0.0:8000/docs

• GET endpoint to get all restaurants:

http://0.0.0.1:8000/restaurant/

• POST endpoint to create a new restaurant:

http://0.0.0.1:8000/restaurant/create

{
    "parameter": {
        "id": "ofdoij509gdpof",
        "rating": 2,
        "name": "MyName",
        "site": "url.com",
        "email": "[email protected]",
        "phone": "5757577987987686",
        "street": "Street",
        "city": "Tuxpan",
        "state": "Jalisco",
        "lat": 17.67684865,
        "lng": -99.6445465
    }
}

• PUT endpoint to update the restaurant info:

http://0.0.0.1:8000/restaurant/update

{
    "parameter": {
        "id": "ofdoij509gdpof",
        "rating": 4,
        "name": "MyName",
        "site": "url.com",
        "email": "[email protected]",
        "phone": "575757798798",
        "street": "Street",
        "city": "Tuxpan",
        "state": "Jalisco",
        "lat": 17.67684865,
        "lng": -99.6445465
    }
}

• DELETE endpoint to remove a restaurant by id:

http://0.0.0.1:8000/restaurant/delete

{
    "parameter": {
        "id": "dsdasd4545"
    }
}

• GET statistics:

http://0.0.0.1:8000/restaurant/statistics/?latitude=17.4341231&longitude=-99.1265732&radius=138

• PostGIS was used to perform the spatial queries. The following query was implemented to count the number of restaurants that fall within a radius in meters.

"""SELECT COUNT(rating) FROM restaurant where ST_Distance('POINT(:lng :lat)'::geography, ST_MakePoint(lng,lat)) * 0.000621371 <= :radius;"""

Where ST_Distance returns the minimum distance between two geometries in miles. - 'POINT(:lng :lat)'::geometry corresponds to the latitude and longitude as parameteres in the GET requests - ST_MakePoint(lng,lat) corresponds to the geometry from the lng and lat values stored in the database

To convert the distance from miles to meters, the conversion factor used was 0.000621371