This is a simple API that returns a UBS (Unidades Básicas de Saúde) list based on a given location (latitude and longitude). It can also import UBSs froma a CSV, through a rake task.

To see a running sample use this link

Health Unit API uses the following technologies:

• Ruby >= 2.5.0
• Rails 5.2.0
• Mongoid (Mongo DB)
• Searchkick (ElasticSearch)

• Thin
• MongoDB
• ElasticSearch
• RSpec (only to test)

• Unicorn

The sample link is runing on AWS using following resources:

• Docker and Docker Compose
• AWS Elastic Container Service (Fargate)
• MongoDB Atlas
• AWS ElasticSearch Service
• AWS Elastic Load Balancer

Health Unit API has one model as MongoDB Document:

You can clone this project from this Github repo

You need have MongoDB and ElasticSearch previouslly installed adn running. Then configure config/mongoid.yml, application.yml with your desired environment settings.

To set up dependencies, in project folder, run:

$ bundle install

It's easy to import ubs.csv (included in this project for practicality). You need run:

$ rails ubs:import

Or

$ rails ubs:import RAILS_ENV=< desired environment >

Once UBS base is dumped, to execute a standalone thin, run:

$ rails s

To test in your browser or preferred client tool, see next item.

Health Unit API has only a valid route. This route return UBSs ordered from closest to farthest UBS

GET '/api/v1/find_ubs'

Sample: //api/v1/find_ubs?query=-6.57,-35.107&page=1&per_page=10

* Requires parameter. See below:

{
    "current_page": 1,
    "per_page": 10,
    "total_entries": 19,
    "entries": [{
        "id": "5c0db2f706518421a78b31f4",
        "name": "UNIDADE BASICA DE SAUDE JOSE BIBI",
        "address": "POVOADO SAO JOAQUIM DOS MELOS",
        "city": "Tuntum",
        "phone": "9935225071",
        "geocode": {
            "lat": -5.89795231819136,
            "lon": -44.8160004615771
        },
        "scores": {
            "size": 1,
            "adaptation_for_seniors": 1,
            "medical_equipment": 1,
            "medicine": 1
        }
    }, {
        "id": "5c0db2e706518421a78b31ec",
        "name": "US OSWALDO DE SOUZA",
        "address": "TV ADALTO BOTELHO",
        "city": "Aracaju",
        "phone": "7931791326",
        "geocode": {
            "lat": -10.9112370014188,
            "lon": -37.0620775222768
        },
        "scores": {
            "size": 2,
            "adaptation_for_seniors": 3,
            "medical_equipment": 1,
            "medicine": 2
        }
    }, {
        "id": "5c0db2f106518421a78b31ef",
        "name": "POSTO DE SAUDE DE BOM JESUS DA ALDEIA",
        "address": "RUA TEOFILO OTONI",
        "city": "Jequitinhonha",
        "phone": "3337411423",
        "geocode": {
            "lat": -16.447874307632,
            "lon": -41.0098600387561
        },
        "scores": {
            "size": 1,
            "adaptation_for_seniors": 2,
            "medical_equipment": 1,
            "medicine": 1
        }
    }, {...}
  ]
}

If query not supplied

{
    "error": {
        "message": "precisa ser informada",
        "full_message": "query precisa ser informada ",
        "field": "query"
    }
}

To execute tests, on project folder run:

$ rspec spec

This project was deployed on AWS ECS Fargate with docker-compose, aws-cli and ecs-cli, previouslly setted up with AWS credentials. Also it was created a cluster, repository for docker container, VPC, subnets, Application Load Balancer and target groups on AWS.

Before push the container it's necessary to set docker-compose.yml and ecs-params.yml with AWS required params

Once all this stuffs be made it's time to deploy. To make it easier, the basic commands to deploy are included on docker-push.sh. It:

• sets your AWS credentials to default cluster config
• builds, tags and pushes container image to AWS Docker repostory
• creates task definitions and update service in the cluster with zero downtime

To execute delpoy, run:

$ ./docker-push.sh

The above stages will be logged and service status will be shown.